mirror of
https://github.com/tiennm99/goclaw.git
synced 2026-10-12 22:14:37 +00:00
dev
63
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 yatulandClaude Opus 5
|
5379cc1163 |
feat(pairing): let an operator make a device pairing permanent
Every approved pairing expires 30 days after approval (pairedDeviceTTL),
and nothing lets an operator change that: owners of a personal setup have
to re-pair their own Telegram account and browser every month. The schema
already treats a NULL paired_devices.expires_at as "never expires" (IsPaired
and the prune both check for it), only no code path ever writes NULL.
- store: SetPairingPermanent(senderID, channel, permanent) on PairingStore,
PG and SQLite. permanent=true clears expires_at, false restarts the
default TTL from now. An already expired pairing is not revived.
PairedDeviceData gains expires_at (Unix ms, null = never), returned by
ApprovePairing and ListPaired.
- gateway: device.pair.approve accepts `permanent`; new admin RPC
device.pair.update {senderId, channel, permanent} for existing pairings,
classified in permissions/policy.go next to the other pairing methods.
- mcp: pairing approve tool accepts `permanent`.
- web UI (Nodes): "Never expires" switch in the approve dialog, an
Expires column, and a Make permanent / Set expiry action per device.
ConfirmDialog takes optional children for the switch.
The default stays 30 days; permanence is an explicit operator choice.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
|
||
|
bilogic
|
72c2c301a5 | get refreshable tokens | ||
|
bilogic
|
c09a72a3bf |
notify channel of pairing approval (#1508)
* notify channel of pairing approval * purge cache when revoked |
||
|
Phan Thong IT
|
7dd0fc2308 |
fix(mcp): stop stripping all-caps business identifiers from tool arguments
isAllCapsPlaceholder() classified ANY string of >=3 chars made only of A-Z and
underscore as an LLM placeholder, and stripEmptyOptionalArgs() then removed it
from the outbound CallTool arguments. Real business identifiers match that shape:
bank codes BKASH, NAGAD, ROCKET, VCB
currency codes BDT, VND, USD, PKR, MMK
enum values SUCCESS, PENDING, ALL
Impact is silent and severe: dropping an optional filter turns a filtered query
into an unfiltered one. The MCP server still answers 200 with the full result set,
so nothing errors — the model reports totals for the entire dataset as if they were
the filtered result, with full confidence.
Observed in production: asking "how many BKASH proxies are active" returned 6,610
(all banks) instead of 1,263. Three different bank codes returned byte-identical
results, which also drove the model into a retry loop until the loop guard fired.
Diagnosis was slow because the removal is invisible: the outbound log records
args_len BEFORE stripping, and stripEmptyOptionalArgs logs nothing at all — while
normalizeArgsForSchema right below it does emit mcp.tool.args.coerced. Confirmed
the argument never reaches the server by echoing the raw JSON-RPC arguments from
RequestContext[CallToolRequestParams] on the server side.
Changes:
- isAllCapsPlaceholder: require multi-word (contains "_") or membership in a known
single-word placeholder list. Keeps the original intent — the doc comment's own
examples (SHOULD_NOT_BE_HERE, DO_NOT_SEND, NOT_APPLICABLE) are all multi-word —
while letting business codes through.
- stripEmptyOptionalArgs: emit slog.Warn("mcp.tool.args.stripped") whenever an
argument is dropped, so this class of bug is one grep away instead of a
multi-hour investigation.
- Tests: guard both layers against regression.
Verified: go build ./... clean; go test ./internal/mcp/ passes.
|
||
|
Duc Nguyenandntduc
|
bb7712a9ff |
fix(collaboration): harden delegated task isolation (#1486)
* feat(collaboration): isolate delegated artifacts and child runs Isolate delegated inputs and outputs behind secure artifact exchange lifecycles. Scope Agent Link tasks by tenant and root agent, and enforce delegation spawn-tree boundaries. Add process-wide child-run admission and preserve logical media paths across native, MCP, and sandbox execution. * fix(collaboration): harden delegated task isolation Enforce tenant and root-agent task scope across migrations and stores. Add exactly-once async completion delivery, delegated sandbox boundaries, and confined artifact and media recovery across runtime surfaces. * fix(collaboration): recover interrupted async tasks * fix(collaboration): normalize persisted child-run status --------- Co-authored-by: ntduc <ntduc@cpp.ai.vn> |
||
|
|
dc605b911e |
feat(mcp): expand CRUD server to near-complete CLI parity (#1440)
* feat(mcp): expand CRUD MCP server to near-complete CLI parity Closes the CLI-vs-MCP tool coverage gap identified by auditing every `goclaw` CLI command against the existing goclaw_* MCP tool set. Adds skill grant/revoke, agent skill pin/unpin, and full CRUD/inspection surfaces for memory, knowledge graph (including dedup/merge/prune), tenants, providers, LLM traces, channel contacts, pending messages, audit activity, system config, tenant storage (list/size/delete/move), scoped agent config export/import, secure-CLI binary registry, and a DB-backed health check. Deliberately out of scope, documented inline where relevant: - `goclaw credentials`: confirmed CLI-local (~/.goclaw/config.yaml + keychain), no server resource to wrap. goclaw_secure_cli_binaries_* covers the closest real, previously-uncovered server resource instead. - Full tar-archive agent export/import (KG + workspace files): the CLI's version streams a multi-section archive with progress events, a shape that doesn't map to a single MCP tool call. Config + context files (the portable "brain") is covered. - `kg extract` (LLM-driven text extraction): goclaw_kg_ingest accepts the same Entity/Relation shapes the extractor produces, so a caller can run extraction itself and hand off the result. Wires 8 new store dependencies (Memory, KnowledgeGraph, Tracing, Contacts, PendingMessages, Activity, SystemConfigs, SecureCLI) through gateway.Server setters -> cmd/gateway.go -> CRUDDeps, following the existing Providers/Tenants pattern. Storage and secure-CLI-binary handlers duplicate internal/http's path-escape/symlink-hiding validation logic (documented inline) since internal/http already imports internal/mcp and the reverse would cycle. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> * fix(mcp): return sessionKey in goclaw_chat_send response for persistent conversations The goclaw_chat_send tool creates a new session internally when sessionKey is empty, but never returned the key to the caller. This prevented using goclaw_chat_history to fetch previous messages in the session. Add SessionKey field to ChatSendResult so callers can: 1. Start a new agent chat without providing sessionKey 2. Receive the sessionKey back in the response 3. Use that sessionKey for follow-up messages and history queries Fixes the training loop pattern: start chat → get sessionKey → call goclaw_chat_history with that key → iterate skill based on actual failures. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> * feat(mcp): add timing diagnostics to goclaw_chat_send for timeout troubleshooting Log request arrival, processing duration, and errors with millisecond precision. Helps identify whether timeouts occur at MCP client→goclaw, goclaw→ollama, or during agent execution. Critical for production debugging. Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com> * fix(gateway): add http.Server timeouts for defensive timeout handling Set explicit timeouts in http.Server: - ReadTimeout: 1h (allow large uploads, long-running agent operations) - WriteTimeout: 1h (allow streaming responses to slow clients) - IdleTimeout: 30s (close idle keep-alive connections quickly) Provides defense-in-depth when Nginx/Traefik timeouts are misconfigured. Matches Nginx timeout (3600s) to prevent race conditions. Timeout chain: traefik 3600s = nginx 3600s = goclaw 3600s Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com> --------- Co-authored-by: Bruno Clermont <bruno.clermont@gmail.com> Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com> |
||
|
|
293aa2a5e1 |
(feat) MCP Server (#1365)
* feat(hooks/observe): add structured output validation hook (ObserveHook) - hooks/types.go: Add ObserveHook interface + BuiltInHookType enum - hooks/dispatcher.go: HookDispatcher emits ObserveHook with PhaseResult payload - pipeline/observe_stage.go: ObserveStage emits hook after ObserveResult built - pipeline/substates.go: ObserveStageResult carries hook results + validation errors - hooks/config.go: BuiltInHookTypeObserve added to BuiltInHookType enum PhaseResult carries structured output, token usage, tool calls + validation errors emitted post-ObserveStage. ObserveHook implementations can validate structured output against schemas, detect tool-call loops, enforce token budgets, etc. Hook fires after ObserveStage produces ObserveResult, before results propagate to next stage. ValidationError returned by hook halts pipeline and propagates error to caller without further stage execution. Co-Authored-By: Claude <noreply@anthropic.com> * ui(hooks): add post_model_response event to web UI - Add event to Zod schema, filter dropdown, and form dialog - Implement conditional test panel UI for model response payload - Add translations (en/zh/vi) for new test panel fields - Updated beta description to reference the new event Co-Authored-By: Claude <noreply@anthropic.com> * feat(mcp): add MCP CRUD server exposing goclaw resources at /api/mcp/ with Bearer token auth and X-GoClaw-Tenant-Id header, default to master tenant * feat(mcp): add goclaw_skills_write_file tool to edit skill files on disk The CRUD MCP server's goclaw_skills_update only touched skill DB metadata, with no way to edit a skill's SKILL.md/file content on the filesystem. Extract the versioned write logic from the web UI's skill file editor (SkillsHandler.handleWriteFile) into skills.WriteVersionedFile so both surfaces share identical validation and versioning, and expose it as a new MCP tool. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> --------- Co-authored-by: Bruno Clermont <bruno.clermont@gmail.com> Co-authored-by: Claude <noreply@anthropic.com> |
||
|
SYNITYandDangTinh311
|
3b8bc1bc4d |
feat(bitrix24): replace two MCP text inputs with a filtered dropdown (#1392)
* feat(bitrix24): replace two MCP text inputs with a filtered dropdown [B24:2794]
Bitrix24 channel creation used to demand two hand-typed strings —
mcp_server_name and mcp_base_url — plus zero indication of which MCP
servers can actually auto-onboard. Typos silently disabled provisioning
and admin had to know which servers implement /api/auto-onboard.
Ship a single dropdown backed by mcp_servers.require_user_credentials,
plus the machinery to make it work end-to-end.
Phase 1 — DB & store
* Promote require_user_credentials from settings JSONB to a top-level
column on mcp_servers (PG migration 000089, SQLite migration v54).
* Backfill from existing settings blobs so no admin needs to re-tick.
* Add MCPServerData.RequireUserCredentials to the Go store layer, plumb
through Create / Get / GetByName / List / Update on both stores,
extend the export DTO, and add require_user_credentials to the HTTP
allowlist.
* Bump RequiredSchemaVersion 87 -> 89 (jumping 88, which was on disk
but not wired) and SchemaVersion 53 -> 54 with an
idempotentColumnMigration guard.
Phase 2 — Bitrix24 channel factory
* Add MCPServerID (UUID string) to bitrix24 InstanceConfig, keep
MCPServerName + MCPBaseURL as legacy fallback with a "deprecated"
doc comment.
* Factory validation accepts either mcp_server_id alone or the legacy
pair; half-config still fails fast.
* initMCPProvisioner prefers GetServer(id) and sources the base URL
from mcp_servers.url when the id path is used. Legacy name path
unchanged so pre-migration configs keep working.
* Log line now carries mcp_server_id + require_user_credentials so
operators can eyeball the wiring.
Phase 3 — Frontend types & MCP form
* Add optional top-level require_user_credentials to MCPServerData /
MCPServerInput in both ui/web and ui/desktop/frontend types.
* mcp-form-dialog reads the top-level flag first and falls back to
settings.require_user_credentials so cached responses from
pre-upgrade backends still render correctly.
* On submit send both the top-level flag AND the legacy settings
entry so mid-rollout backends stay consistent.
Phase 4 — Bitrix24 channel form dropdown
* New mcp-select field type + MCPServerSelect component. Uses the
shared useMCP() react-query cache and filters client-side to
servers whose require_user_credentials is true (OR settings
JSONB during the migration window).
* Explicit "None (disable MCP provisioning)" option so admins can
clear the binding without editing config JSON.
* Legacy mcp_server_name / mcp_base_url text inputs kept in the
Advanced panel, relabelled "(legacy)" with pointer help text.
Phase 5 — channel_instances.config backfill
* PG migration 000090 and SQLite migration v55 rewrite existing
bitrix24 channel_instances.config to add mcp_server_id by
resolving mcp_server_name against mcp_servers, tenant-scoped
via agents.tenant_id (channel_instances doesn't carry tenant_id
directly).
* Idempotent — only touches rows already carrying
mcp_server_name that lack mcp_server_id. Legacy keys are left
in place so provisioner.go can still fall back for unmigrated
or future-created legacy configs.
* down.sql drops the mcp_server_id key. Provisioner immediately
reverts to the legacy fallback path.
Tests
* provisioner_test.go: three new cases exercise the mcp_server_id
path (invalid UUID string, valid UUID with missing row, valid
UUID with a per-user row). fakeMCPStore gains a serversByID
map and a real GetServer implementation.
* Existing legacy-config tests unchanged and still green.
Verification
* go build ./... && go build -tags sqliteonly ./...
* go vet ./internal/mcp/... ./internal/channels/bitrix24/...
./internal/store/... ./internal/http/...
* go test ./internal/mcp/... ./internal/channels/bitrix24/... -> ok
* Live-tested against a local docker image on the goclaw-deploy
postgres. Migrations 89 + 90 applied cleanly. Three existing
bitrix24 channels (bitrix-sales / nguyen-dao-openline / tieu-vi)
had their configs backfilled with the b24-syn-mcp UUID and the
provisioner boots with require_user_credentials=true. UI dropdown
correctly shows only b24-syn-mcp (the only server with the flag
ticked) alongside a "None" clearer option.
Surface parity
* Gateway server: store + factory + provisioner + HTTP allowlist.
* API contract: adds require_user_credentials + mcp_server_id
as optional fields on existing routes. No new endpoints.
* Web UI: MCP form + Bitrix24 channel form + shared types.
* CLI/runtime package: N/A because no CLI subcommand reads the
mcp_server_id field.
* fix(bitrix24): derive auto-onboard base URL from mcp_servers.url origin [B24:2794]
The Phase 2 refactor swapped provisioner base-URL sourcing from the
legacy per-channel MCPBaseURL config field (which historically stored the
MCP server's ORIGIN, e.g. https://mcp.example.com) to mcp_servers.url,
which stores the JSON-RPC ENDPOINT the agent loop dials (e.g.
https://mcp.example.com/mcp). The two are semantically different but
share a single column.
mcp_client.newMCPClient then appends "/api/auto-onboard" to whatever
baseURL it receives, so the id-path started POSTing to
".../mcp/api/auto-onboard" — 404 for every per-user credential mint and
refresh. Existing users kept working only until their cached access
tokens expired.
Fix: derive the origin (scheme://host[:port]) from server.URL before
handing it to the auto-onboard client. The legacy path is untouched
because MCPBaseURL from channel config is already the origin.
https://b24-mcp-dev.synity.so/mcp -> https://b24-mcp-dev.synity.so
https://mcp.example.com/mcp/ -> https://mcp.example.com
https://mcp.example.com -> https://mcp.example.com
Table-driven test covers six shapes plus four error cases (empty,
whitespace-only, no scheme, no host). Updated the existing
TestInitMCPProvisioner_MCPServerID fixture to seed a URL with the /mcp
subpath so it regression-guards the same code path.
Verified live: user 614 sent a message that triggered the expired-cred
refresh branch; goclaw logged "self-refreshed user credentials
created=false" and the agent immediately reported
"mcp.user_tools_loaded user=614 tools=2". Before this fix the same event
logged 'auto-onboard failed: mcp auto-onboard: 404 Not Found'.
Surface parity:
- Gateway server: provisioner + one new helper (deriveAutoOnboardBaseURL).
- API contract: N/A because the wire shape hasn't changed.
- Web UI: N/A because the UI still writes mcp_server_id verbatim.
- CLI/runtime: N/A because no CLI reads the derived base URL.
---------
Co-authored-by: DangTinh311 <dangtinh31193@gmail.com>
|
||
|
MToan
|
09d8660f3b |
fix(mcp): derive claude_cli bridge tool surface from agent policy (#1377)
The bridge exposed a static BridgeToolNames subset that drifted from the tool registry: use_skill, datetime, knowledge_graph_search and skill_manage were never added, while the system prompt's skill-loading protocol requires agents to call use_skill. claude_cli agents following the protocol hit a nonexistent tool and could fabricate results. Implement the structural fix recommended in #1373 triage: - register the full bridge-capable surface (registry minus hard exclusions spawn/create_forum_topic) instead of the static list - gate BOTH tools/list (new WithToolFilter) and tools/call through one shared predicate bridgeToolAllowed: * callers WITH a verified agent policy get exactly the policy-filtered surface (same WouldAllow check the call path always enforced) * callers WITHOUT one (anonymous, or agent without tools_config) keep the legacy conservative BridgeToolNames set - no exposure widening - downgrade the per-call denial log Warn->Info; list filtering makes probes of denied tools rare and the call gate is the intended enforcement point Fixes #1373 |
||
|
|
a3fe5f7858 |
fix(mcp): resolve tool descriptions/schemas consistently across all MCP tool-listing paths (#1369)
* fix(mcp): return bare tool names with descriptions for already-connected MCP servers, fixing empty descriptions/schemas in agent prompts * fix(mcp): resolve live tool descriptions/schemas in ListToolsForAgent for legacy prefixed tool_allow entries The previous fix (d612da8e) only corrected the admin "browse/select tools" endpoint. The actual runtime path building the LLM system prompt's tool list, Manager.ListToolsForAgent, still looked up tool_cache/registry entries keyed by the raw stored tool_allow name. Agents with grants captured before that fix have tool_allow persisted as registered (prefixed) names instead of bare MCP tool names, so every lookup missed and every tool in the prompt preview came back with an empty description and schema. Add bareMCPToolName to normalize both legacy-prefixed and current bare tool_allow/tool_deny entries before matching, and resolveMCPToolInfo to prefer the live registry's current description/schema (falling back to the settings tool_cache) instead of trusting whatever was persisted verbatim. Both the explicit-allow and cache-enumeration branches now share this single resolution path. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> --------- Co-authored-by: Bruno Clermont <bruno.clermont@gmail.com> Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com> |
||
|
|
90e7035100 |
fix(mcp): tenant isolation, tool policy engine, and prompt preview fixes (#1333)
* fix(prompt): remove duplicate Team Members section from system prompt The TEAM.md context file already provides the Members section with better formatting. The system prompt section was redundant and inconsistently formatted. - Remove buildTeamMembersSection() call from system prompt building - Remove unused buildTeamMembersSection() function * fix(mcp): wire store to manager for prompt preview tool visibility MCP manager needs database access to query configured servers for tool visibility in prompt preview. - Add SetStore() method to MCP manager - Wire pgStores.MCP to manager after initialization - Add debug logging for MCP initialization flow * fix(systemprompt): hide Tooling section when agent has no tools The Tooling section header and boilerplate were displayed even when the agent had no tools available. Add early return in buildToolingSection() to skip the section entirely when toolNames is empty. This reduces prompt noise for agents with no tool access. * feat(mcp): cache tool descriptions for prompt preview visibility MCP tools now show descriptions in prompt preview without requiring a live server connection. - Add CacheToolDescriptions() method to MCPServerStore (PG + SQLite) - Cache tool descriptions in settings['tool_cache'] when server connects - Use cached descriptions in ListToolsForAgent (fallback: hints → cache → global) - Descriptions are auto-populated from live server manifest on connection - Admins can still override via tool_hints in server settings * test(mcp): add CacheToolDescriptions to MCPServerStore test fakes Commit 5ce410b6 added CacheToolDescriptions() to the MCPServerStore interface but missed updating test mock implementations, breaking go vet across internal/mcp, internal/agent, internal/http, and internal/channels/bitrix24. Add no-op implementations matching each fake's existing style. * fix(providers): enforce per-agent tool policy for Claude CLI provider The Claude CLI provider (stdio+MCP bridge) was not enforcing per-agent tool policy, unlike other providers where the policy-filtered tool list already drives the system prompt's Tooling section. Two gaps closed: 1. --disallowedTools was previously skipped entirely when no MCP config path was resolved, letting the CLI subprocess run with its full native toolset (Bash, Edit, Read, Write, Glob, Grep, WebFetch, WebSearch) regardless of agent policy. It's now unconditional and derived from the agent's actual allowed-tools list (state.Tool.AllowedTools), mapped to Claude CLI's native tool names. 2. The MCP bridge server executed any tool call without checking the calling agent's policy. It now resolves the agent's policy from context (via the existing HMAC-verified agent lookup) and denies calls to tools outside that agent's allowed set, logging security.mcp_bridge_denied on denial. Both gaps were closed using existing plumbing (PolicyEngine.WouldAllow, AgentData.ParseToolsConfig, the bridge context middleware) — no new cross-cutting mechanism was introduced. * feat(web): show MCP/tool schemas in system prompt preview dialog The prompt-preview API response includes a separate `tools` field (the actual JSON schemas sent to the LLM as the tools API parameter) alongside `prompt` (the system prompt text), but the web UI only rendered `prompt`, silently dropping the tools list. Add a collapsible Tools section to both the full-screen System Prompt dialog and the inline agent-detail preview, showing tool count, name, description, and expandable parameter schema per tool. i18n keys added to en/vi/zh locales. * fix(prompt): render pinned skills on bootstrap turns Pinned skills are documented (web UI copy) as "always inlined in the system prompt", but the entire Skills section was gated behind !cfg.IsBootstrap, so pinned skill XML never appeared on bootstrap turns (first message of a session) despite the promise. Separate pinned-skill rendering from bootstrap-suppressed guidance: - Bootstrap + pinned skills present: render pinned XML only, no search/manage guidance (which stays suppressed as before) - Non-bootstrap: unchanged behavior - Minimal/none modes: pinned skills always render regardless of bootstrap state Add regression tests covering all four prompt modes on bootstrap turns, plus a non-bootstrap guard confirming existing behavior is preserved. * fix(skills): resolve managed skills directory per-tenant, not master-only skills.Loader was wired at startup to scan a single fixed directory (the master tenant's managed-skills dir), making any skill belonging to a non-master tenant invisible to both pinned-skills prompt resolution and skill_search/use_skill, regardless of DB visibility settings. - Loader now resolves the calling tenant's managed-skills directory per-call via context (store.TenantIDFromContext), never enumerating other tenants' directories - Skill cache is now tenant-keyed to prevent slug collisions and cross-tenant cache leaks across tenants using the same skill slug - gateway_setup.go passes the root data dir instead of a pre-resolved master-tenant path Write-side tooling (skill_manage, publish_skill) was already correctly tenant-scoped per-operation — no changes needed there. Added TestLoader_ManagedSkills_TenantIsolation proving two tenants with same-slug/different-content skills never see each other's content, including after cache population from a different tenant's lookup. Known follow-up (not in this commit): skill_search's BM25 index is still a single process-global index shared across tenants, which is a related but separate cross-tenant search-result leak requiring its own scoped fix (per-tenant index maps + threading tenant context through ensureIndex/rebuildIndex). * fix(tools): scope skill_search BM25 index per-tenant SkillSearchTool held a single process-global BM25 index built once from whichever tenant's context first triggered ensureIndex, then reused for all subsequent Execute() calls regardless of caller — leaking one tenant's skill search results into another's, the search-path counterpart to the managed-directory bug fixed in 7b4668ad. - index/lastVersion are now keyed per-tenant (map[uuid.UUID]*tenantIndexState) - ensureIndex resolves the calling tenant from context and only builds/reads that tenant's index entry, never touching another tenant's cached state - Builtin/bundled skills remain visible in every tenant's index (Loader already merges those tiers correctly per 7b4668ad) Loader.Version() remains a single global counter — a version bump in one tenant causes unnecessary rebuilds in others but does not cause cross-tenant leakage, an acceptable tradeoff to avoid scope creep. Added TestSkillSearchTool_TenantIsolation proving two tenants with same-slug/different-content skills never see each other's search results, including after cache population from a different tenant. * fix(tools): fix group-spec expansion in tool policy engine PolicyEngine.registry was only ever set via SetRegistry(), which was never called in production (only in one test) — so pe.registry was permanently nil in production. Every group-expansion helper (applyProfile, intersectWithSpec, unionWithSpec, subtractSpec, expandSpec, matchDenySpec, filterByCapability) silently dropped any "group:*" spec entry instead of expanding it when registry was nil. Concretely: "group:mcp" (auto-injected into agentToolPolicy.AlsoAllow for any agent with MCP tools) never resolved to real tool names, so MCP tools connected successfully and appeared in prompt text (which reads the registry directly, bypassing PolicyEngine) but were never included in the actual ChatRequest.Tools payload sent to the LLM — confirmed live via mcp.agent.tools_loaded tools=6 immediately followed by mcp.filtered_tools mcp_defs_count=0 in the same request. This affects any agent relying on group-based grants, not just MCP. PolicyEngine is a shared/global singleton used concurrently across all agents (constructed once at gateway startup), so mutating a registry field per-call would be a data race. Fix instead threads the registry as an explicit parameter from FilterTools down through all internal group-expansion helpers, and adds IsDenied/WouldAllow registry parameters, removing the dead SetRegistry() mechanism entirely. Also fixes group expansion for the per-user-MCP-tools path: FilterTools is sometimes called with a userToolOverlay wrapping a *Registry rather than a *Registry directly; added Unwrap() to userToolOverlay so the new registry-resolution logic works for both cases. Added 4 tests proving group expansion works via the threaded parameter alone (no SetRegistry): plain registry allow, userToolOverlay allow, deny-side group expansion, and the WouldAllow bridge-server path. Blast radius note: this restores intended access for every agent configured with group:* specs (group:mcp, group:vault, group:goclaw, group:coding, etc.) that were silently inert before. Existing agent configs relying on group grants will gain the tool access they were nominally already configured for. * fix(mcp): cache tool descriptions from pool-connected servers too 5ce410b6 added tool-description caching (for prompt-preview visibility) only inside connectServer. connectViaPool — the separate connect path used when MCP connections go through the shared pool — never got the same caching hook, even though it shares the same underlying connectAndDiscover wire handshake. Confirmed live: cloudflare/docker connected via connectViaPool and received real descriptions over the wire, but prompt preview still showed blank descriptions because this path never wrote to the cache. Also removes the temporary mcp.connect.raw_tool debug log added earlier this session for diagnosing the same issue — no longer needed now that the root cause is fixed. * chore(skills): remove temporary pinned-skills diagnostic logging Confirmed live: pinned skills (caveman, infra-ansible-knowledge) now resolve correctly end-to-end for tenant-scoped agents. Debug logging added to trace the resolution chain is no longer needed. * fix(tools): deny always wins over AlsoAllow group grants AlsoAllow's unionWithSpec could reintroduce a tool explicitly listed in Deny, since it added tools back from allTools without re-checking deny specs. Previously masked because AlsoAllow's group-expansion was also broken (fixed in f7af95de this session) — group specs silently expanded to nothing, so this ordering bug never manifested. Now that group expansion works, an admin-denied tool that's also reachable via a group:* AlsoAllow entry (e.g. group:mcp) would silently reappear. Re-apply deny-spec subtraction as a final step after AlsoAllow union, for both global and per-agent policy, so deny always wins regardless of which allow mechanism tries to add a tool back. Added tests proving global and per-agent Deny correctly override an overlapping AlsoAllow group grant, while sibling non-denied tools in the same group remain allowed. * fix(mcp): enumerate cached tools instead of wildcard placeholder in prompt preview ListToolsForAgent (the prompt-preview path) collapsed any server with an empty ToolAllow (unrestricted grant — the common case) into a single "server__*" placeholder entry, even when tool_cache already had every real tool name and description from connect time (5ce410b6, 8ffd67b9). This meant agents with unrestricted MCP server access never saw individual tool names or descriptions in prompt preview, forcing trial-and-error tool usage. When ToolAllow is empty and tool_cache is populated, enumerate every cached tool (skipping any explicitly denied) and emit one MCPToolPreviewInfo per tool, matching the construction logic already used for the ToolAllow-non-empty case. Falls back to the single placeholder only when tool_cache is also empty (server never connected). The live (non-preview) conversation path, buildMCPToolDescs, does not have this bug — it resolves tool identity from the live connected registry, never from ToolAllow, so no placeholder shortcut exists there. Added tests covering both the cache-populated enumeration case and the no-cache placeholder fallback. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> * fix(prompt): filter alias re-injection and add MCP schemas to preview ToolDefs Prompt-preview's tools: schema array (PreviewResult.ToolDefs) had two bugs, both preview-only — confirmed live conversations already use correctly-filtered tool payloads via buildToolsPayload/PolicyEngine.FilterTools, unaffected by either bug: 1. Alias re-injection iterated ALL registry aliases globally with no check against the deny-filtered toolNames list, letting denied tools reappear via their alias name (e.g. a denied canonical tool still showing up as its Claude-Code-compat alias like Bash/Edit/Write/Read). Now skips any alias whose canonical tool isn't in the filtered set. 2. MCP tool descriptions (from the store-based, connection-free ListToolsForAgent path) only ever fed the prompt TEXT section, never got converted into ToolDefinition schema objects — so the tools: array never showed MCP tools at all, even when the MCP section of the prompt text correctly listed them. Now appends a ToolDefinition per MCP tool with name+description populated and a placeholder {"type":"object"} Parameters schema, documented as preview-only (real parameter schemas require a live MCP connection, only available during an actual conversation turn). Added tests proving denied-tool-alias exclusion and MCP tool inclusion in preview ToolDefs. * feat(skills): inline full content for pinned skills instead of pointer-only Web UI documented "Pinned skills are always inlined in the system prompt", but BuildPinnedSummary was just BuildSummary with an allowlist filter — same as the general searchable skill list: name/description(truncated)/location pointer only, requiring use_skill+read_file round trips to get actual content. No code path inlined real SKILL.md content for pinned skills specifically. BuildPinnedSummary now reads and inlines full SKILL.md content (frontmatter stripped) per pinned skill inside <skill_instructions> tags. Per-skill (10000 bytes) and total (30000 bytes) size caps fall back to the original pointer-only format with a note when a skill is too large to inline, so oversized skills degrade gracefully instead of blowing the prompt budget. Added tests: full-content inlining, size-cap fallback, and tenant isolation for the new inline path (mirroring the existing managed- skills tenant isolation test). * fix(mcp): real parameter schemas and full policy enforcement in prompt preview Three interconnected fixes to prompt preview, none affecting the live conversation path (which was already correct): 1. Real MCP parameter schemas instead of a useless empty placeholder. tool_cache previously stored only name+description; extended to also capture the real JSON Schema from the MCP server's tools/list response (CachedToolInfo{Description, Parameters}) at connect time, for both direct-connect and pool-connect paths. Preview now shows complete, real input schemas instead of {"type":"object"} with no properties — verified against actual wire data from a live cloudflare MCP server showing genuinely rich schemas (zone_id, type, name, content, ttl, proxied, all typed with descriptions and correct required arrays) that were previously being discarded. Backward-compatible: old-shape cache entries degrade gracefully to description-only rather than crashing, self-healing on next connect. 2. Global tool deny now enforced in preview. BuildPreviewPrompt previously hand-rolled a partial policy reimplementation (per-agent deny only, explicitly skipping the full PolicyEngine "because runtime state isn't available in preview") — but PolicyEngine.WouldAllow already handles this per-tool-name without needing channel context. A tool denied via the global config (not per-agent) would appear in preview despite being correctly denied in every real conversation. Preview now calls WouldAllow per candidate tool, with a graceful per-agent-deny-only fallback when no PolicyEngine is wired (e.g. in tests). 3. MCP tools now also subject to the same policy check — previously the MCP tool supplement (store-based, connection-free tool listing) added MCP tool names unconditionally, bypassing WouldAllow entirely, so a denied MCP tool could still appear in preview. Added tests for all three: real-schema presence, global-deny exclusion for both core and MCP tools, and backward-compat cache handling. * fix(prompt): remove redundant per-tool MCP enumeration from prompt text Now that MCP tool schemas in the tools: API parameter are real and complete (1290d4f1), the ## MCP Tools prompt-text section's per-tool "- mcp_x__y: description" enumeration is pure duplication with zero added value — the model already gets each tool's real schema (including description) via tools:. buildMCPToolsInlineSection now keeps only the behavioral instructions that aren't expressible via JSON schema and thus aren't duplicated: prefer-MCP-over-core-tools guidance, and the optional-parameter guidance (don't guess/fill optional fields). The per-tool name+ description enumeration loop is removed. Section still only appears when the agent has MCP tools (len(cfg.MCPToolDescs) > 0, unchanged gate). Updated tests to assert the enumeration is gone while the behavioral instructions remain; ToolDefs assertions are now the authoritative check for MCP tool allow/deny filtering behavior (prompt text no longer enumerates names at all). * test(agent): update TeamContextInjection test for removed Team Members section TestBuildSystemPrompt_TeamContextInjection asserted the presence of a 'Team Members' prompt-text section that was intentionally removed in 71d33180 (duplicate of the canonical TEAM.md-context-file Members section, which has better formatting). The test was never updated to match, causing it to fail on every run since. Moved the assertion from wantIn to wantNotIn for the 3 affected subtests -- BuildSystemPrompt correctly no longer renders team-member roster info directly; that info now comes exclusively from the TEAM.md context-file mechanism, outside this test's isolated scope. * fix(prompt): resolve real registry for WouldAllow calls in preview BuildPreviewPrompt's two WouldAllow calls hardcoded reg=nil, silently breaking group:* expansion (e.g. group:mcp) needed to resolve the AlsoAllow grant production actually uses to grant MCP tool access (resolver_helpers.go's agentToolPolicyWithMCP injects AlsoAllow: ["group:mcp"]). With reg=nil, WouldAllow could match literal tool names fine (the 18 core/static tools) but could never resolve group-based grants, so every MCP tool silently failed WouldAllow and was excluded from preview -- confirmed live via curl: 18 tools returned, zero mcp_* ones, for an agent with genuinely working MCP access in real conversations. Live conversations were never affected -- internal/mcp/bridge_server.go's WouldAllow call already correctly passes a real registry. Fix resolves a real *tools.Registry from deps.ToolLister via tools.ResolveConcreteRegistry (the same helper used at the live call site), passing it to both WouldAllow calls instead of nil. Falls back to nil gracefully for test mocks that don't implement the full ToolExecutor interface, preserving existing test behavior. Added a test proving an MCP tool granted via the exact production AlsoAllow: ["group:mcp"] pattern is now correctly included in preview ToolDefs, where the old reg=nil bug would have silently excluded it. * fix(prompt): use literal deny check for MCP tools in preview, not group expansion The MCP-tools policy gate added in 1290d4f1 called WouldAllow with a real registry (per 9b5fd2eb), which requires group:mcp expansion against that registry to grant access via the production AlsoAllow: ["group:mcp"] pattern. But MCP tools are only ever registered into ephemeral per-agent registry clones at live connection time (manager_connect.go) -- never into the shared/global registry preview uses. group:mcp always resolved empty in preview's connection-free context, so WouldAllow denied every MCP tool -- confirmed live via tool-name diff: live conversations correctly included all 6 MCP tools, preview included zero. MCP access-granting is already correctly handled by ListToolsForAgent's own per-server tool_allow/tool_deny grant logic (confirmed working correctly earlier this session). The preview gate only needs to catch the narrower case of a literally-denied tool name via global/per-agent policy config -- it never needed group expansion. Replaced WouldAllow with IsDenied(nil, name, agentPolicy), which forces a pure literal-name match with zero registry dependency, matching the existing usage pattern already established elsewhere in policy.go. This class of bug cannot recur: there's no registry-passing code path left in this check to silently reintroduce group-expansion dependence. Added a test proving MCP tool inclusion in preview is independent of group-expansion outcome (no AlsoAllow: group:mcp needed for a non-denied tool to appear). Also reverts the temporary loop.filtered_tool_names/ preview_prompt.filtered_tool_names diagnostic logging used to capture the live-vs-preview tool-name comparison that diagnosed this bug. * fix(http): preserve real MCP parameter schemas through HTTP preview adapter mcpPreviewAdapter.ListToolsForAgent (the HTTP-layer glue converting mcp.MCPToolPreviewInfo to agent.MCPToolPreviewInfo for BuildPreviewPrompt) only copied RegisteredName and Description, silently dropping Parameters -- a bug present since this adapter was introduced (2499d0be/7e250244), unrelated to today's other MCP preview fixes. This was masked until d5fc6344 fixed MCP tools being excluded from preview entirely (a separate bug) -- once MCP tools started appearing again, this pre-existing adapter gap became visible: tools showed up correctly, but always with the bare {"type":"object"} placeholder instead of their real cached schema (confirmed live: update_dns_record missing its 7 real properties). One-line fix: copy Parameters through in the adapter's struct literal. Added a regression test constructing a real *mcp.Manager with populated tool_cache, asserting the adapter's output preserves specific real schema properties (not just non-nil Parameters) -- verified this test fails without the fix and passes with it. --------- Co-authored-by: Bruno Clermont <bruno.clermont@gmail.com> Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com> |
||
|
Bruno ClermontandBruno Clermont
|
2499d0be78 |
fix(mcp): show configured MCP tools in prompt preview (#1321)
* fix(mcp): show configured MCP tools in prompt preview MCP tools were only visible in prompts if currently loaded in registry (from active sessions). Prompt preview showed empty MCP tools section. Solution: Query MCP store for configured servers and tool lists. Show tools that would be available, not just currently loaded. - internal/mcp/manager.go: add ListToolsForAgent() for store-based tool discovery - internal/agent/preview_prompt.go: supplement live registry with store tools - internal/http/agents.go: add mcpPreviewMgr field and setter - internal/http/agents_prompt_preview.go: create adapter bridging mcp.Manager to preview - cmd/gateway.go: wire up MCPPreviewAdapter during setup * fix(mcp): load MCP servers from database at startup, not config file Single source of truth: MCP servers are now loaded from mcp_servers table at gateway startup, instead of from the config file. This ensures that MCP servers configured via web UI are automatically available without requiring config file updates. - cmd/gateway_setup.go: remove config fallback, add initMCPFromDB() helper - cmd/gateway.go: call initMCPFromDB() after store initialization - internal/mcp/manager.go: add SetConfigs() for runtime configuration Now mcpMgr will be non-nil when MCP servers are in the database, enabling MCP tools in prompt preview. --------- Co-authored-by: Bruno Clermont <bruno.clermont@gmail.com> |
||
|
Duc Nguyenandntduc
|
ac2c382fbd |
fix(mcp): normalize stringified container arguments (#1319)
Co-authored-by: ntduc <ntduc@cpp.ai.vn> |
||
|
mozaa
|
42a20c0f22 |
chore(mcp): allow mcp-server-darwin-arm64 in validation allowlist (#1107)
Local pkg-helper installs an arm64 mcp-server binary on macOS hosts; the existing allowlist (deno, bun, uvx, ...) didn't cover it so MCP server config validation rejected legitimate entries. |
||
|
thotam
|
0ae55991bb |
feat(mcp): MCP OAuth 2.1 client for tool servers (#1196)
* feat(mcp): MCP OAuth 2.1 client — full implementation with tests
Implements a complete MCP OAuth 2.1 authorization flow for tool servers that
require user-delegated access, covering all layers from DB to UI.
- discovery.go: RFC 9728 protected-resource → RFC 8414 AS metadata → OIDC
fallback chain with 5-min in-memory cache and InvalidateCache()
- dcr.go: RFC 7591 Dynamic Client Registration with response size guard
- flow.go: PKCE (S256) authorization code flow — StartFlow(), ExchangeCode(),
ClientCredentials(), auto-cleanup of expired flows; carries AS issuer through
PendingFlow for status display
- refresher.go: OAuthTokenProvider with in-memory token cache, automatic refresh
on expiry, per-user vs global slot isolation, InvalidateCache/InvalidateServer
- migrations/000074 + SQLite schema: mcp_oauth_tokens with AES-256-GCM encrypted
access/refresh tokens, partial unique index for global vs per-user rows,
ON DELETE CASCADE from mcp_servers
- store.MCPOAuthTokenStore: Upsert, Get/GetUser, Delete/DeleteUser, and
DeleteServerOAuthTokens (purge all rows for a server)
- PostgreSQL + SQLite implementations
- POST /v1/mcp/oauth/start — discovery + optional DCR + PKCE redirect URL;
client_credentials completes server-side (no redirect) and returns completed=true
- GET /v1/mcp/oauth/callback — exchange code, persist token, publish WS event;
payload built via json.Marshal (no reflected XSS via error_description)
- GET /v1/mcp/oauth/status/{id}, DELETE /v1/mcp/oauth/token/{id} — admin-gated
- POST /v1/mcp/oauth/discover/{id} — on-demand discovery probe
- All outbound calls go through the SSRF-safe client with pinned IPs
- pkg/protocol/mcp_events.go: EventMCPOAuthComplete routed only to the initiating
user (admins in-tenant included); fail-closed across tenants
- getUserMCPTools() injects Authorization: Bearer from OAuthTokenProvider; on a
401 for OAuth servers it purges the cached token so the next turn re-resolves
- handleUpdateServer purges all OAuth tokens (global + per-user), drops the
refresher cache, and evicts the pool when a server's URL or OAuth config
(client_id / endpoints / grant_type / scope / auth_type) changes — so the
status UI and agent never use a token minted for the old resource/AS
- MCPOAuthDialog (WS-driven), unified user-credentials dialog, OAuth settings
fields; handles the no-redirect client_credentials completion
- internal/mcp/oauth/*_test.go: discovery cache, PKCE, DCR, refresher
- internal/http/mcp_oauth_test.go + mcp_update_oauth_purge_test.go: routes, auth
gating, WS event, purge-on-URL/OAuth-config-change
- tests/integration: store + encryption + tenant isolation, E2E start→callback,
DeleteServerOAuthTokens
- internal/gateway/event_filter_test.go, internal/agent/loop_mcp_user_test.go
* fix(mcp): return 400 on OAuth callback with code but missing state
The callback handler rendered a 200 HTML page whenever code or state was
absent. An auth code WITH a missing state is a malformed / CSRF-risk
callback (state is the CSRF token), so reject that case with HTTP 400.
A bare hit with neither code nor state (user opening the URL directly),
provider errors, and exchange failures keep their 200 HTML popup page.
Adds a status code parameter to writeCallbackHTML. Fixes the
TestOAuthCallbackMissingState integration regression while keeping
TestHandleCallbackMissingCodeAndState (no params -> 200) green.
* fix(mcp): scope-based OAuth auth + honor manual OAuth endpoints
Addresses the two MCP/OAuth security-review findings.
Finding 1 — authorization. mcp_oauth_tokens is tenant-scoped, but
start/status/revoke were gated only by requireAuth(RoleAdmin), an RBAC
role check, not tenant membership, so a RoleAdmin caller could act on a
tenant they don't administer. A blanket requireTenantAdmin would have
broken per-user self-service, which the UI exposes (the per-user
MCPUserCredentialsDialog shows an "Authorize" button to regular users for
their own credentials). Instead mirror the existing per-user MCP
credentials model (resolveTargetUserID in mcp_user_credentials.go):
- start/status/revoke accept any authenticated user; each handler calls
authorizeOAuthScope.
- a caller may manage their OWN per-user token (self-service); the
global/server token (user_id="") and other users' tokens require
tenant-admin (owner bypass), so a RoleAdmin that is not a tenant admin
is rejected.
- discover stays admin-only (it only previews AS metadata for a server).
Add a TenantStore dependency. Tests cover self-service, on-behalf-of-
another (403), and global-by-non-tenant-admin (403).
Finding 2 — honor manual OAuth config end-to-end. The UI sent use_dcr /
auth_endpoint / token_endpoint and the update path fingerprinted them for
purge, but handleStart always discovered + DCR'd and ignored them. Now:
- use_dcr=false (a *bool, so legacy/absent stays discover+DCR) skips
discovery/registration and uses the operator endpoints, SSRF-validated.
- token_endpoint is always required; auth_endpoint only for auth-code
grants — client_credentials needs no authorization URL, matching the UI
which hides that field for that grant.
- the refresher already refreshes against the stored token_endpoint and
the callback persists it, so manual-mode tokens refresh correctly.
- oauthFingerprint includes use_dcr (nil normalized to true) so toggling
DCR mode purges stale tokens.
- the web form only serializes manual endpoints when use_dcr is off.
Audited all MCP dialogs (form, global OAuth, per-user credentials, grants,
tools): OAuth dialogs handle completed/auth_url identically and read
config from stored server settings; runtime connect uses the stored token
via the refresher (no re-discovery).
Tests: manual auth-code + client_credentials endpoints, missing/SSRF
endpoints, and the full self/global/on-behalf authorization matrix.
|
||
|
Zezae Oh
|
c31dd21934 |
feat(mcp): opt-in allowlist for trusted private MCP hosts (#1248)
Registering a remote MCP server (sse/streamable-http) whose hostname resolves to a private IP is rejected at config-validation time by the SSRF guard, with no production escape hatch -- the only bypass is a test-only loopback flag. Self-hosted MCP servers on a private network are therefore unregisterable, even though the runtime MCP client connects to them fine (Test Connection succeeds and lists tools; only create/update input validation blocks). Add an opt-in, operator-configured allowlist (GOCLAW_MCP_ALLOWED_HOSTS, empty by default) of trusted hostnames exempt from the private/loopback IP block during MCP server URL validation only: - security.ValidateAllowingHosts(url, allowedHosts): like Validate but skips the private/loopback block for allowlisted hostnames. The cloud-metadata/link-local (169.254/fe80), multicast and unspecified ranges are never exempted, even for allowlisted hosts. - mcp.SetAllowedHosts wires the operator allowlist into ValidateURL / ValidateServerConfig; default empty => no behavior change. - web_fetch / webhook / redirect SSRF paths are unchanged (those stay agent-influenced and fully guarded). Matching is case-insensitive on the pre-resolution hostname. |
||
|
Zezae Oh
|
719d5934bb |
fix(mcp): expose delegate tool via MCP bridge (#1247)
The BridgeToolNames allowlist omitted "delegate", so claude-cli / Claude-Code-backed agents never received the tool even when their orchestration mode resolved to "delegate" with active agent_links. The native agent loop injects delegate per orchestration mode, but bridge sessions only see this curated allowlist, making inter-agent delegation impossible from bridge-backed agents. delegate is safe to expose unconditionally: it self-gates via CanDelegate/agent_links and resolves its source agent from the X-Agent-ID header context, exactly like team_tasks (already bridged). |
||
|
Duy Nguyen
|
f8875e46d6 | fix: reject path-based MCP runtime commands | ||
|
Duy Nguyen
|
878c33e964 | fix: harden MCP stdio validation | ||
|
Goon
|
a63080714a |
Merge remote-tracking branch 'upstream/dev' into codex/merge-nextlevelbuilder-goclaw
# Conflicts: # internal/pipeline/think_stage.go # internal/pipeline/tool_stage.go # tests/integration/git_adapter_ssh_test.go |
||
|
viettranx
|
f8f910e540 | fix(mcp): keep grant checker cache invalidation active | ||
|
Duy /zuey/
|
2cd750ac44 |
feat(channels): add context capability admin surface
Squash merge PR #115 after resolving changelog and SQLite migration-map conflicts with current dev. Renumbered channel-context PostgreSQL migration to 000075 and bumped PG required schema to 75 plus SQLite schema to 44 so it follows the run timeline migration. Local checks passed: go test ./..., go build ./..., go build -tags sqliteonly ./..., go vet ./..., and pnpm -C ui/web build. PR CI run 26705617311 passed release-versioning, go, and web. |
||
|
Duy /zuey/
|
bc3bc25c98 |
fix(mcp): eliminate spurious grant-revoked errors (tool filter + session reset + system-user bypass) (#87)
* feat(mcp): filter tools at registration + detect FastMCP session reset with force-reconnect Two foundational MCP reliability improvements: (1) Tool allow/deny filtering at BridgeTool registration: Previously the runtime grant-check at execute time surfaced "grant revoked" errors when the LLM called a tool it wasn't allowed to call. Filter upfront in both the per-agent registration path (manager_connect.go) and per-user registration path (loop_mcp_user.go). Adds tool_filter.go with IsToolAllowed + tests. This eliminates the "registered then runtime-denied" loop. (2) FastMCP session reset detection + force-reconnect: FastMCP/Python mcp servers reject tools/call as "invalid during session initialization" when the upstream session lifecycle resets but our pool still holds the old Mcp-Session-Id. Detector matches three known phrasings (FastMCP, mcp-go, mcp-go transport ErrSessionTerminated) in session_reset.go. On detection, BridgeTool.Execute requests a force-reconnect via atomic CAS dedup so N concurrent failing calls collapse to one reconnect. Health loops skip ping while pending so a server answering ping in "initializing" state cannot clobber connected=true before the fresh Initialize completes. Includes 30s timeout, structured slog telemetry, concurrent CAS dedup test. Files: tool_filter.go + tool_filter_test.go (new), session_reset.go + session_reset_test.go (new), manager_connect.go (connectServer/connectViaPool signatures + registerBridgeTools/registerPoolBridgeTools filter logic + reconnPending skip), manager.go (connectAndFilter + connectServer call signature changes), loop_mcp_user.go (filter-at-register block + WithForceReconnect wiring), pool.go (reconnPending skip), bridge_tool.go (session reset detection + WithForceReconnect callback). * fix(mcp): self-heal grant cache + bypass per-user grant for system/empty userID Two production fixes for "MCP tool: grant revoked" recurring on song-nhi-v2. (1) System-user bypass in ListAccessible: Registration uses LoadForAgent(ctx, agentID, "") while execute uses IsAllowed(ctx, agentID, "system", ...). The LEFT JOIN on mcp_user_grants could match a stale disabled row keyed user_id='system' and silently filter the server out only at execute. Skip the join entirely for synthetic owner identities (userID="" or "system") so registration and execute see the same set. Applied to both PostgreSQL (mcp_servers_access.go) and SQLite (mcp_servers_access.go). (2) Grant-checker no-cache on empty allowByServer: grant_checker.loadEntry now skips the cache write when allowByServer is empty. Without this, a single transient empty result pinned permanent denial until a bus invalidate fired. Re-queries until the empty condition clears, then caches normally. Includes TestStoreGrantChecker_EmptyEntryNotCached. |
||
|
Duy /zuey/
|
ef829c1ab4 |
fix(bitrix24): resolve PR 8-10 stack findings
Resolve the PR #8/#9/#10 stack on current dev, including Bitrix24 install callback hardening, migration renumbering, duplicate-domain fail-closed routing, UI textarea/mobile cleanup, and review hardening. |
||
|
Duy Nguyen
|
5a189330f9 | fix(test): support windows test execution | ||
|
viettranx
|
c81cf6cf39 |
feat(security): MCP validation + hooks context isolation
Phase 02 - MCP input validation: - Add command allowlist (node, python, npx, uvx, etc.) - Block dangerous args (--eval, -e, -c, exec(), etc.) - SSRF protection via security.Validate() with DNS rebinding check - Fail-closed env var allowlist for headers - Integrate validation in create/update/import handlers Phase 03 - Hooks async context leak: - Use context.WithoutCancel to preserve TenantID/UserID in async hooks - Increase audit write timeout 2s → 5s - Increase circuit breaker store timeout 1s → 2s Also: Add security guard tests for skills upload |
||
|
viettranx
|
8b8da3a3dd |
fix(mcp): unified grant revocation with per-agent tool group isolation
- Add GrantChecker interface with cache + event-bus invalidation for revalidating MCP tool grants at execution time (not just dispatch) - Move toolGroups from package-level global to per-Registry field; Clone() gives each agent Loop isolated tool groups, eliminating cross-agent race condition - BridgeTool.Execute rechecks grant validity before executing MCP methods; uses serverID field for cache key - Wire MCPGrantChecker through ResolverDeps to MCP Manager and LoopConfig - Add integration test for grant revocation scenarios - Update policy functions to accept *Registry parameter |
||
|
|
ad893908a5 |
fix(telegram): propagate local_key for forum topic routing in team notifications (#800)
* fix(telegram): propagate local_key for forum topic routing in team notifications
Team task status messages (dispatched, completed, progress) were always
delivered to the General topic in Telegram forum groups because the
notification pipeline had no access to the originating topic's local_key.
Root cause: wireTeamProgressNotifySubscriber in gateway_events.go published
OutboundMessage with no Metadata, so the Telegram adapter had no
message_thread_id to route to the correct forum topic.
Fix has two parts:
1. Team notify path (root cause):
- Add LocalKey field to TeamTaskEventPayload (protocol)
- Extract local_key from tool context in WithContextInfo()
- Add LocalKey to NotifyRoutingMeta
- Pass LocalKey through to OutboundMessage Metadata in both
leader mode (InboundMessage) and direct mode (OutboundMessage)
2. MCP bridge context (supporting):
- Propagate local_key and session_key through bridge HTTP headers
- Add X-Local-Key and X-Session-Key to BridgeContext
- Extract and inject into tool context in gateway middleware
- Include in HMAC signature for integrity
Closes #798
* fix(telegram): pass LocalKey from task metadata in all dispatch/fail broadcast sites
The initial fix added LocalKey to the event payload and WithContextInfo(),
but 4 broadcast call sites use individual With* options instead of
WithContextInfo — so LocalKey was never populated for:
- fallback_dispatch (team_tasks_create.go)
- dispatch_unblocked (team_tool_dispatch.go)
- post_turn dispatch (team_tool_validation.go)
- blocker/fail (team_tasks_blocker.go)
Add WithLocalKey() option function and extract TaskMetaLocalKey from task
metadata at each site, matching the existing TaskMetaPeerKind pattern.
* test(mcp): add HMAC verification tests for extra params (localKey, sessionKey)
- Add tests for SignBridgeContext/VerifyBridgeContext with extra params
- Test backward compat fallback for pre-localKey sessions
- Test that param order matters in signature
- Add clarifying comment for routing context injection security model
---------
Co-authored-by: Jens Henke <jens@henke.dk>
Co-authored-by: viettranx <viettranx@gmail.com>
|
||
|
viettranx
|
a8f0f05600 |
test(config,skills,backup,mcp): wave 2 integration tests
Add Wave 2 package coverage: - config/config_extras_test.go: NormalizeAgentID, ExpandHome, ResolveAgent, env overlays - skills/search_bm25_test.go: BM25 index/search/ranking - skills/loader_test.go: SKILL.md loader, frontmatter, dedup, versioning - backup/fs_archive_test.go: ArchiveDirectory, skip fn, tar prefix - backup/manifest_dsn_test.go: SanitizeDSN, ParseDSN, WritePgpass, CleanEnv - mcp/util_bm25_test.go: pool, manager, bridge BM25, env resolution |
||
|
viettranx
|
de6bdf2290 |
refactor(mcp): DRY reconnect logic, fix client pointer thread safety
- Extract reconnectWithBackoff() shared by Manager and Pool reconnect paths - Use clientPtr.Load() in Stop/Evict/evict paths to avoid race with fullReconnect - Add tool staleness note to fullReconnect - Remove dead poolEntry.Client() |
||
|
Kai (Tam Nhu) Tran
|
86ef70906a |
fix(mcp): full reconnect for SSE/HTTP after server-side restart (#812)
* fix(mcp): full reconnect for SSE/HTTP after server-side restart When an MCP server using SSE or streamable-http transport restarts (container redeploy, crash, OOM), the old client holds a stale session ID. tryReconnect only called Ping() on the dead client, which keeps POSTing to /message?sessionId=<dead> — the server returns 404, and after maxReconnectAttempts the connection is permanently dead. For pool connections, poolHealthLoop only set connected=false without attempting any reconnect at all. Fix: - Store connection params (url, headers, command, args, env) in serverState so tryReconnect can create fresh connections. - tryReconnect now has two phases: fast path (ping existing client for transient blips) and slow path (close old client, create fresh one via createClient + Start + Initialize, swap ss.client). - Add updateBridgeToolClients to propagate the new client pointer to all registered BridgeTools after a full reconnect. - Add poolTryReconnect with the same two-phase pattern for pool-managed connections. - Add BridgeTool.swapClient for safe client pointer replacement. Closes #810 * fix(mcp): address review findings — atomic client pointer, close-after-swap, cooldown Red-team review found critical issues in the initial reconnect fix: 1. Data race: BridgeTool.client written by healthLoop, read by Execute concurrently without synchronization. Fix: BridgeTool now holds *atomic.Pointer[mcpclient.Client] shared with serverState.clientPtr. Execute uses atomic Load(), reconnect uses atomic Store(). 2. Close-before-swap: old client was closed before new one was ready. If createClient failed, ss.client pointed to closed client permanently. Fix: create and validate new client first, swap atomically, then close old. 3. Pool BridgeTool orphaning: poolTryReconnect swapped ss.client but existing BridgeTools held old pointer with no update mechanism. Fix: atomic.Pointer propagates automatically — no explicit update needed. 4. Permanent death: after maxReconnectAttempts, server was permanently dead with no recovery. Fix: 5-minute cooldown then reset attempts. 5. healthFailures not reset in fast path: minor inconsistency fixed. Also fixes external caller in loop_mcp_user.go and adds ClientPtr() accessor to poolEntry for atomic pointer sharing. * refactor(mcp): extract fullReconnect helper, add nil guard, clarify dual-pointer design - Extract shared fullReconnect() used by both tryReconnect and poolTryReconnect — eliminates 30-line duplication (-17 lines net) - Add nil guard on clientPtr.Load() in BridgeTool.Execute to prevent nil deref in edge cases (test fixtures, initialization race) - Add doc comment on serverState explaining dual-pointer design (client for healthLoop, clientPtr for BridgeTools) |
||
|
Luan VuandLuvu182
|
0e2282be8f |
fix: use errors.Is() for sentinel comparisons + remove unused @xyflow/react (#727)
Replace direct sentinel error comparisons (==, !=) with errors.Is()
across the codebase. Direct comparison breaks when errors are wrapped
with fmt.Errorf("...: %w", err), which can cause missed matches for
sql.ErrNoRows, io.EOF, context.DeadlineExceeded, and context.Canceled.
Affected packages:
- store/pg: sql.ErrNoRows in secure_cli, agents_export_team_*, mcp_export_queries
- store/sqlitestore: sql.ErrNoRows in schema migration
- tools: context.DeadlineExceeded in shell, credentialed_exec
- mcp: context.DeadlineExceeded in bridge_tool
- providers: context.Canceled in acp_provider, anthropic_stream_test
- updater: io.EOF in tar extraction
Also removes unused @xyflow/react dependency from web UI — it was
replaced by react-force-graph-2d but never cleaned up from package.json.
Co-authored-by: Luvu182 <208665161+Luvu182@users.noreply.github.com>
|
||
|
viettranx
|
4c485d6fa1 |
fix(mcp): retry stdio transport initialization with exponential backoff
Heavy MCP servers (FastMCP with 80+ tools, OAuth servers) take 3-5s to start their stdin read loop. Without retries, connectAndDiscover sends the initialize request immediately after fork/exec, gets EOF, and permanently fails with "transport closed". Add a retry loop (4 attempts, 2s/4s/8s backoff) for stdio transports only. SSE/HTTP transports fail immediately since connection-refused is definitive. The retry respects context cancellation to avoid blocking shutdown. Closes #385 |
||
|
viettranx
|
ff4d370d27 |
fix(mcp): require 3 consecutive ping failures before marking server disconnected
Single transient errors (e.g. 504 from upstream proxy) no longer instantly disconnect MCP servers. Requires healthFailThreshold (3) consecutive failures before setting connected=false and triggering reconnect. Applies to both Manager.healthLoop and poolHealthLoop. |
||
|
viettranx
|
e48ba1df7f |
fix(mcp): prevent LLM hallucination of optional tool parameters
3-layer defense against GPT-5.4 filling all optional MCP tool params with fabricated values (e.g. api_key:"optional", proxyUrl:"http://example.com"): Layer 1 — bridge_tool.go: expand placeholder detection to catch "optional", "skip", example URLs; type-aware empty string handling (keep for string-typed, strip for non-string); add propertyType() helper. Layer 2 — schema_strict.go: OpenAI strict mode transform — optional props become nullable unions, all props required, additionalProperties:false. Constrained decoding prevents invalid output. Only enabled for first-party OpenAI/Codex providers. Layer 3 — systemprompt_sections.go: concrete WRONG/RIGHT examples in MCP optional param instruction. |
||
|
viettranx
|
f623ef9d55 |
feat(mcp): hybrid search mode — keep first 40 tools inline, defer rest
Instead of all-or-nothing when MCP tool count exceeds threshold, keep first 40 tools registered inline and only defer the excess to BM25 search via mcp_tool_search. System prompt now shows both inline descriptions and search guidance in hybrid mode. Also raises skill inline count from 40 to 60 (token limit is the real bottleneck for skills). |
||
|
viettranx
|
bdfc0fadfb |
fix(config): MCP env: resolution, channel field filter, orphan provider event, workspace fallback (#348, #297, #295, #431)
- manager.go: add resolveEnvVars() to expand env:VARNAME in MCP headers (copy-safe)
- manager_env_test.go: 3 test cases for env resolution
- validate.go: add "name" to channelInstanceAllowedFields
- validate_test.go: test channel instance name field retention
- bus/types.go: add TopicAgentDeleted event + AgentDeletedPayload
- agents_delete.go: emit agent:deleted event with provider name for async cleanup
- gateway.go: log-only subscriber for orphaned provider warning (no auto-delete — FK safety)
- resolver.go: fallback to deps.Workspace for master tenant agents with empty ag.Workspace
(fixes filesystem escape where filepath.Join("","system") resolved to /system)
|
||
|
viettranx
|
21b6c454ca |
feat: merge pipeline, per-user credentials, unified picker, group contacts
- Enable merge UI for linking channel contacts to tenant_users - Contact → tenant_user resolution with cached lookup (60s TTL) - MCP per-user credentials via user-keyed connection pool - Secure CLI per-user credentials with AES-256-GCM encryption - Unified UserPickerCombobox searching contacts + tenant_users - Group contact collection with chat title in all channels - Group permission inheritance via wildcard user_id="*" - Fix heartbeat using wrong userID in group chats - Filter internal senders from contact collection - Add contact_type column (user/group) to channel_contacts - SQLite schema v2 migration for desktop edition |
||
|
viettranx
|
f8278cfaff |
fix(mcp): strip placeholder values from optional tool args and guide LLMs on optional params
LLMs often send empty strings, "null", "SHOULD_NOT_BE_HERE" etc. for optional MCP tool parameters instead of omitting them, causing servers to reject invalid values (e.g. UUID validation fails on ""). - Add stripEmptyOptionalArgs() to clean optional args before MCP calls - Detect known placeholders (null/none/nil) and all-caps patterns - Add system prompt guidance for optional parameter handling - Empty string "" is intentionally NOT stripped (may be valid for text fields) |
||
|
viettranx
|
78253f1841 |
refactor: decompose consumer, agent loop, team tools, and MCP manager
6-phase pure refactor — zero behavior change, all tests pass. Phase 1: DRY event payloads + metadata keys - BuildTaskEventPayload() helper with functional options replaces 24 inline constructions - 26 metadata string constants replace ~66 magic strings across dispatch/consumer Phase 2: Consumer handler decomposition - ConsumerDeps struct collapses 11-13 positional params to 3 per handler - Extract startTaskLockRenewal() and resolveTeamTaskOutcome() Phase 3: loop.go extraction (978→686 LOC) - enrichInputMedia(): media persistence, ref collection, tag enrichment - injectTeamTaskReminders(): leader/member task context injection - buildFilteredTools(): policy, bootstrap, channel, iteration filtering - collectRefsByKind() helper eliminates 3x copy-paste Phase 4: team_tasks_mutations split (630→318 LOC) - executeCreate extracted to team_tasks_create.go - resolveTeamAndTask() helper used by 13 execute* functions Phase 5: MCP LoadForAgent split (~90→~15 LOC body) - resolveServerCredentials() + connectAndFilter() helpers Phase 6: Comments + DI - 5 inaccurate comments fixed - ToolExecutor interface for dependency inversion (Loop.tools field) - FilterTools accepts ToolExecutor instead of concrete *Registry |
||
|
Kai (Tam Nhu) Tran
|
30708ae79d |
feat(providers): support Codex OAuth pools with inherited routing defaults
* feat(auth): support named chatgpt oauth providers - add provider-scoped ChatGPT OAuth routes and CLI support - persist refresh tokens per provider and reject provider-type collisions - wire provider OAuth setup flows in the dashboard and setup UI Refs #448 * feat(agent): add chatgpt oauth account routing - add agent other_config routing for manual and round-robin selection - reuse routed provider resolution across resolver and pending loaders - add router, parser, and agent advanced dialog coverage for multi-account use Refs #448 * docs(api): describe chatgpt oauth routing - document named-provider ChatGPT OAuth auth routes - describe agent-side account routing and round-robin behavior - update OpenAPI agent config schema and provider type enum Refs #448 * fix(store): add missing agent key context helpers * feat(ui): clarify chatgpt oauth account setup and routing * docs(providers): align chatgpt oauth alias examples * feat(agent): add codex pool activity dashboard * fix(providers): harden codex oauth alias setup * feat(codex-pool): improve routing dashboard UX - redesign the Codex/OpenAI pool page around saved-pool checkpoints and live evidence - add clearer selection, attention, and recent-proof states for pool members - make the lower panels fill the remaining desktop viewport while staying responsive * fix(store): resolve context helper merge duplication * feat(oauth): add codex pool quota and observation APIs - add quota inspection and observation endpoints for ChatGPT Subscription (OAuth) providers - teach codex routing to surface pool activity, observation metadata, and quota-aware readiness - extend tests and HTTP docs/OpenAPI for the new pool monitoring flows * feat(web): add codex pool quota monitor and controls - add provider quota fetching, readiness badges, and live routing evidence on the account pool page - redesign pool setup and activity panels for multi-account management with localized copy updates - keep the live monitor internally scrollable and compact the account cards for better viewport fit * fix(web): clarify pool routing labels - rename the recent request badge from Direct to Selected - restore compact quota bars in the live pool cards * feat(codex-pool): add runtime health dashboard - derive per-provider success and failure health from routed Codex traces - surface routing, quota, and recent request evidence in the pool UI - align provider alias guidance and owner access with the dashboard role model * docs(auth): document tenant scoping and key roles * fix(auth): harden tenant and codex pool access control * fix(providers): align codex pool runtime defaults * feat(ui): tighten codex pool responsive layout * feat(chatgpt-oauth): refine codex pool management UX * feat(chatgpt-oauth): surface quota bars on provider pages - add compact quota bars to Codex provider rows and provider detail - fetch quota only for ready visible provider rows and ready detail aliases - fix managed-member detail visibility and tighten provider locale copy |
||
|
Viet Tran
|
cd022699f6 |
feat: multi-tenant isolation — complete implementation (#359)
* feat(security): multi-tenant user data isolation (Plan 1)
Comprehensive user data isolation for non-owner system users:
- API key identity binding: owner_id column forces user_id on auth,
prevents spoofing via X-GoClaw-User-Id header
- Sessions: ownership checks on list/preview/patch/delete/reset,
non-admin users see only their own sessions
- Cron: user_id filtering on list, ownership checks on mutations
- Server-side WS event filtering: agent/chat/session/cron/team events
scoped per-user instead of broadcast to all clients
- Web UI role guards: RequireAdmin on 15 admin-only pages, role
propagated from WS connect response to auth store
- Tracing/activity: user_id enforcement for non-admin HTTP callers
- Teams: HasTeamAccess membership checks on get/delete/list
- Skills: fail-closed ownership check (deny non-admin if store
doesn't support owner lookup)
- HTTP auth: requireAuthBearer now enforces owner_id + user context
for file/media downloads (was missing)
- Dead code: removed delegation_history, handoff_routes tables and
all related handlers/store code
- New: team_user_grants table for user-to-team access control
Migration 000026: api_keys.owner_id + team_user_grants + DROP legacy tables
* feat(security): multi-tenant foundation — tenants table, tenant_id propagation, permission cache (Plan 2)
Add tenant isolation infrastructure across the entire gateway:
Schema (migration 000027):
- Create tenants + tenant_users tables with master tenant seed
- Add tenant_id column to 30 user-scoped tables (NOT NULL DEFAULT master)
- api_keys.tenant_id nullable (NULL = system-level cross-tenant key)
- Create builtin_tool_tenant_configs + skill_tenant_configs for per-tenant overrides
- Drop custom_tools table (agent loop integration never wired)
Store layer:
- TenantStore interface + PGTenantStore (CRUD tenants + tenant_users)
- TenantID field on AgentData + APIKeyData
- tenant_id in agents/api_keys/skills SQL (Create, Get, List)
Context propagation:
- WithTenantID/TenantIDFromContext (uuid.Nil = fail-closed)
- WithCrossTenant/IsCrossTenant (owner/system admin flag)
Auth tenant resolution:
- HTTP: resolveAuthBearer sets TenantID/CrossTenant on all 5 auth paths
- WS: handleConnect sets tenantID/crossTenant on Client
- API key 2-tier: NULL = cross-tenant (system), set = tenant-scoped
Runtime isolation:
- Event bus: TenantID field on Event, fail-closed filter in event_filter.go
- Cron: tenant context injected in RunJob handler
- Subagent: tenant validation prevents cross-tenant spawn
- Security logging: tenant_id in auth resolution logs
Tenant management:
- WS RPC: 7 methods (tenants.list/get/create/update, tenants.users.*)
- HTTP: 7 endpoints (/v1/tenants/*)
- Slug validation + path traversal prevention
- Role validation (owner/admin/operator/member/viewer)
Infrastructure:
- PermissionCache: 4 sub-caches (tenant resolve, role, agent access, team access)
- tenant_paths.go: filesystem path helpers with master-tenant backward compat
- i18n: MsgInvalidRole key + translations (en/vi/zh)
Dead code removed: custom_tools store, HTTP handler, DynamicToolLoader (-828 lines)
* feat(security): tenant query filtering + workspace isolation (Plan 3)
Add WHERE tenant_id filtering to all 30+ tenant-scoped store queries,
wire workspace filesystem isolation, and harden restrict_to_workspace.
Store query filtering:
- Add tenantClauseN/tenantIDForInsert/requireTenantID helpers
- Filter all SELECT/INSERT/UPDATE/DELETE by tenant_id for non-cross-tenant
- Refactor SessionStore.GetOrCreate and CronStore.AddJob/ListJobs to
accept context.Context for tenant propagation
- System skills (is_system=true) bypass tenant filter for all tenants
- Special cases: GetByKey (channels), GetByHash (auth) skip filter
Workspace isolation:
- Resolver computes tenant-scoped workspace + dataDir for non-master tenants
- Add WithTenantSlug/TenantSlugFromContext to context propagation
- Add TenantStore + Workspace to ResolverDeps
- Force effectiveRestrict() to always return true (multi-tenant security)
- Remove restrict_to_workspace from agentAllowedFields
UI cleanup:
- Remove custom-tools pages, types, routes, constants (backend removed in Plan 2)
- Clean tool-name-select component of custom tools references
* feat(security): session ctx propagation + execMapUpdate tenant guard (Plan 4)
Session store:
- Add ctx to AddMessage, SetSessionMetadata, SetAgentInfo, List, Save
- List now filters by tenant_id for non-cross-tenant callers
- Save uses ExecContext for cancellation support
- All ~15 callers updated to pass ctx
execMapUpdate tenant guard:
- Remove deleted_at IS NULL from execMapUpdateWhereTenant (only agents has soft-delete)
- Migrate 8 callers to execMapUpdateWhereTenant: agent_links, channel_instances,
mcp_servers, secure_cli, tracing, teams, skills_crud, cron_update
- Add ctx to UpdateSkill, UpdateJob interfaces + all callers
Deferred: cron scheduler global cache (correct by design — system process),
browser per-tenant isolation (separate plan).
* refactor(store): add context.Context to all SessionStore interface methods
Complete ctx propagation across all 24 SessionStore methods for:
- Future tenant-aware DB operations
- Request cancellation/timeout support
- Distributed tracing capability
Updated ~15 files including all callers in agent loop, gateway methods,
heartbeat ticker, tools, and CLI commands.
* fix(security): remove context.Background() shadowing in gateway handlers
Critical fix from code review: gateway agent handlers (create, update,
delete, identity, files, links, teams) were creating ctx := context.Background()
which shadowed the handler's ctx that carries tenant_id. This breaks
tenant-scoped agent queries for non-master tenants.
- Remove ctx shadowing in 7 agent handler files
- Add ctx param to resolveAgentUUID/resolveAgentInfo helpers
- Use store.WithCrossTenant in resolver (system-level operation)
* feat(security): tenant-scoped UNIQUE constraints for multi-tenant isolation
Update UNIQUE indexes to include tenant_id, allowing same names across tenants:
- agents: (agent_key) → (tenant_id, agent_key) WHERE deleted_at IS NULL
- sessions: (session_key) → (tenant_id, session_key)
- skills: (slug) → (tenant_id, slug)
- mcp_servers: (name) → (tenant_id, name)
- channel_contacts: (channel_type, sender_id) → (tenant_id, channel_type, sender_id)
Code changes:
- GetByKey now filters by tenant_id (same pattern as GetByID)
- ON CONFLICT clauses updated for sessions and skills
- Channel consumer uses WithCrossTenant for agent resolution
- Down migration restores original constraints
* fix(security): close remaining tenant isolation gaps from final audit
Critical fixes:
- gateway_setup: WithCrossTenant for default agent lookup at startup (C6)
- channel_contacts: ON CONFLICT updated to (tenant_id, channel_type, sender_id) (Q15)
- agents.Delete: tenant filter on DELETE (Q1)
High priority fixes:
- agents: List, GetDefault, ShareAgent, RevokeShare, ListShares, CanAccess,
ListAccessible, Update unset-default — all now tenant-scoped
- skills_crud: DeleteSkill now takes ctx, verifies tenant ownership
- mcp_servers, channel_instances, secure_cli: Delete methods tenant-scoped
- WithCrossTenant added to: gateway team notifications, team_tool_cache,
pending_messages GetDefault
* fix(migration): add tenant_id to usage_snapshots unique index
Update idx_usage_snapshots_unique to include tenant_id, preventing
cross-tenant upsert collisions when different tenants have agents
with same provider/model/channel combination.
* feat(security): cron tenant guard + browser per-tenant isolation
Phase 3 — Cron API tenant guard:
- Add ctx to 5 CronStore methods (GetJob, RemoveJob, EnableJob, RunJob, GetRunLog)
- All API-facing cron ops now filter by tenant_id (prevents cross-tenant CRUD)
- RemoveJob/EnableJob return "not found" on tenant mismatch (no enumeration)
- GetRunLog JOINs cron_jobs for tenant filtering
- UpdateJob internal reads scoped by tenant (defense-in-depth)
- Scheduler-internal methods (GetDueJobs, refreshJobCache) unchanged (system-level)
Phase 4 — Browser per-tenant isolation:
- Per-tenant incognito browser contexts via rod Incognito() (separate cookie jars)
- All page access (Snapshot, Screenshot, Navigate, Click, Type, etc.) validated
via getPageForTenant — blocks cross-tenant access by targetID
- OpenTab creates pages in tenant's incognito context
- ListTabs scoped to tenant's incognito context
- ConsoleMessages validates page ownership
- Stop/reconnect properly cleans up incognito contexts
* feat(security): isolation gaps + per-tenant config (Plan 5)
Part A — Isolation Gap Fixes:
- Merge migration 028 into 027: add tenant_id to llm_providers +
config_secrets, fix UNIQUE constraints for paired_devices +
channel_instances
- providers.go: tenant filtering on all CRUD queries
- config_secrets.go: ON CONFLICT (key, tenant_id)
- pairing_store: add ctx to all 7 interface methods, remove hardcoded
MasterTenantID, update ~15 channel caller files
- Session cache: prefix keys with tenantID to prevent cross-tenant
collision. DB queries (loadFromDB, Save, Delete, LastUsedChannel)
add tenant filter
- config_permissions cache: prefix keys with tenantID
- Cron ListJobs: fail-closed when tenant context missing
Part B — Per-Tenant Configuration:
- Provider Registry: compound key tenantID/name with fallback to
master tenant. GetForTenant/ListForTenant/RegisterForTenant
- Resolver: uses tenant-aware provider lookup + disabled tools query
- Agent loop: filter disabled tools from LLM tool definitions
- Builtin tool tenant configs: store interface + PG implementation +
PUT/DELETE HTTP endpoints
- Skill tenant configs: store interface + PG + ListAccessible LEFT
JOIN to exclude disabled skills per tenant
- OAuth: DBTokenSource with tenantID field for tenant-scoped token
refresh
- All HTTP provider handlers use RegisterForTenant/UnregisterForTenant
* feat(security): channel tenant propagation + MCP per-user credentials (Plan 6)
- Propagate tenant_id from channel_instances through BaseChannel →
InboundMessage → agent loop context (fixes 5-point break in tenant flow)
- Inject tenant context in WS router dispatch for all gateway methods
- Add MCP per-user credential overrides (api_key, headers, env) with
AES-256-GCM encryption and HTTP API endpoints
- Rewrite MCP pool with tenant-scoped keys, slot semaphore, idle eviction,
and credential rotation support (Evict per tenant+server)
- Bypass pool for users with custom credentials (separate connections)
- Fix MCP APIKey never passed to connections (inject as Authorization header)
* fix(security): close remaining tenant isolation gaps from Plan 1-6 audit
- Add tenant_id to 6 missing tables: agent_context_files,
skill_agent_grants, mcp_agent_grants, team_tasks, spans,
embedding_cache (migration 027)
- Fix tid==uuid.Nil fallback to fail-closed (return error) in 8 update
methods: agent_links, teams, skills, channel_instances, secure_cli,
cron, mcp_servers, tracing
- Add tenant filter to bare DELETEs: DeleteLink, DeleteTeam
- Add tenant filter to queries: ListChildTraces, GetMonthlyAgentCost,
CountAgentGrantsByServer, ListAccessible (MCP), ReviewRequest,
ResolveGroupTitles, buildTraceWhere
- Fix missing tenant_id in INSERTs: CreateSkill, GrantToUser,
ReviewRequest grant INSERTs
- Add tenant filter to api_keys: List, Revoke, Delete
- Fix cron scanJob/RemoveJob/EnableJob fallthrough patterns
* fix(security): inject tenant context into channel handler entry points
Channel handlers used context.Background() which lost tenant context,
causing store operations to either fail-closed or default to master
tenant. Now all 10 handler entry points inject tenant from BaseChannel.
* fix(security): tenant filters for teams, tasks, skills (Plan 6b audit)
- Teams: add tenant filter to GetTeamForAgent, ListMembers,
ListIdleMembers, KnownUserIDs (JOIN agent_teams for tenant check)
- Teams: add tenant_id to GrantTeamAccess INSERT, tenant filter to
RevokeTeamAccess, ListTeamGrants, HasTeamAccess
- Team tasks: add tenant_id to CreateTask INSERT, fail-closed
UpdateTask, tenant filter on all 7 query/delete methods
- Skills: add tenant filter to RevokeFromAgent, ListAgentGrants
- Skills: add ctx param + tenant filter to ToggleSkill
- History: annotate context.Background() locations with TODOs for
future tenant injection (requires PendingHistory struct refactor)
* fix(security): add tenant_id to 4 missing team tables + fix INSERTs
Add tenant_id column to: agent_team_members, team_task_comments,
team_task_events, team_task_attachments (migration 027).
Fix INSERT statements to include tenant_id: AddMember,
AddTaskComment, RecordTaskEvent, AttachFileToTask.
* fix(migration): cast UUID literals in tenant_users seed + usage_snapshots index
PostgreSQL doesn't auto-cast string to UUID in SELECT and expression
index contexts. Add explicit ::uuid casts to prevent migration failure.
* docs: add multi-tenant architecture guide for integrators
Comprehensive solution doc covering auth model, WS protocol, event
system, data isolation, API reference, and integration patterns.
Target audience: developers building custom frontends or SaaS on GoClaw.
* feat(ui): multi-tenant awareness + tenant admin page (Plan 7)
Backend:
- Enrich WS connect response with tenant_name, tenant_slug, cross_tenant
- Add tenants.mine WS method (any user, returns own memberships)
- Parse tenant_hint in connect params for browser pairing multi-tenant
- Wire tenantStore to MethodRouter for connect-time tenant lookup
Frontend:
- Auth store: tenantId, tenantName, tenantSlug, isCrossTenant, availableTenants
- WS client: capture tenant fields from connect, send tenant_hint
- WS provider: auto-fetch tenants.mine on connect
- useTenants() shared hook for all tenant-aware components
- Tenant indicator in sidebar connection status
- Tenant admin page (/admin/tenants) with list + create dialog
- Tenants nav in sidebar (cross-tenant admin only)
- i18n: tenants namespace (en/vi/zh)
- Type updates: tenant_id on AgentData, ApiKeyData
* refactor(ui): move tenant selector into user menu dropdown in topbar
Replace simple logout button with a Radix Popover user menu showing:
- User ID display
- Tenant selector (when multi-tenant: list all tenants with check mark)
- Logout button
Remove tenant indicator from connection-status.tsx (now in topbar).
Tenant switch saves slug to localStorage and reloads for reconnect.
* feat(ui): add logout confirmation dialog
Show destructive confirm dialog before logout via ConfirmDialog
component. Added logoutConfirm i18n key for en/vi/zh.
* fix(ui): security hardening — hide admin nav, fix route guard, fix refresh
- Hide System nav group for non-admin roles in sidebar (was visible to all)
- Replace RequireAdmin with RequireCrossTenant guard on /admin/tenants route
- Add RequireCrossTenant component to require-role.tsx
- Fix refresh button animation: use isFetching instead of isLoading
- Clean up connection-status.tsx (remove tenant indicator, now in topbar)
* feat: cross-tenant admin tenant scope selector
Backend: add tenant_scope connect param. Cross-tenant clients can
narrow their scope to a specific tenant (slug). applyTenantScope()
sets client.tenantID and clears crossTenant flag.
UI: user menu shows "All Tenants" option for cross-tenant admins.
Selecting a tenant saves slug to localStorage as tenant_scope,
reload reconnects with narrowed scope. "All Tenants" clears scope.
* feat: provisioning API key scope + tenant detail page (Plan 8)
Backend:
- Add operator.provision scope for limited tenant management
- Add HasScope() method to gateway Client
- Allow provision-scoped keys to create tenants + add users
- Allow provision-scoped keys to create tenant-bound API keys
Frontend:
- Tenant detail page with user management (list, add, remove)
- Clickable tenant list rows navigate to detail
- i18n: tenant detail keys (en/vi/zh)
- Route /admin/tenants/:id with RequireCrossTenant guard
* fix: tenant scope keeps admin privileges + UI pattern fixes
Backend:
- applyTenantScope keeps crossTenant=true (retains admin features)
- Router: scoped cross-tenant injects WithTenantID (filters data)
while keeping admin role for method access
UI:
- Fix "All Tenants" check mark (compare against nil UUID string)
- Fix tenant label when scope active (show selected tenant name)
- Use ConfirmDialog for user removal (was hand-rolled)
- Add DialogDescription to add-user dialog (Radix a11y)
- Fix table min-w-[600px] consistency
- Fix column header mismatch (was "role", should be "created")
* fix(ui): clean up tenant detail header — remove redundant info panel
Remove duplicate slug/status/created panel. Info now shown in
PageHeader description (slug + date). Status badge removed (redundant
with description). Cleaner, consistent with other admin pages.
* fix(ui): redesign tenant detail with info cards + user cards
* feat(ui): tenant selection gate — require tenant before app access
- Add tenantSelected flag to auth store (persisted via localStorage)
- WS provider auto-selects: single-tenant user auto, cross-tenant
admin defaults to "All Tenants", zero-tenant user blocked
- RequireAuth gate: redirect to /select-tenant when connected but
no tenant selected
- New TenantSelectorPage: centered card layout matching login page,
"All Tenants" amber card for cross-tenant admin, per-tenant cards
with role badges, no-access state with logout button
- i18n: selectTenant, noAccess keys (en/vi/zh)
* fix(security): scope events for cross-tenant admin with tenant_scope
Event filter was checking !crossTenant before filtering — scoped
cross-tenant admins (crossTenant=true + tenantID set) bypassed
tenant event filtering. Now checks tenantID != Nil regardless of
crossTenant flag, ensuring scoped admins only see their chosen
tenant's events.
* fix(security): HTTP API now respects tenant_scope for gateway token
Root cause: UI uses HTTP API (/v1/agents, /v1/mcp/servers, etc.)
for data fetching. HTTP auth middleware with gateway token always
set CrossTenant=true with no tenant filtering. tenant_scope only
worked for WS connection, not HTTP requests.
Fix:
- HTTP client sends X-GoClaw-Tenant-Scope header from localStorage
- HTTP auth resolves header slug → tenant UUID via tenantStore
- requireAuth: CrossTenant + TenantID → WithTenantID (scoped)
- Wire InitTenantStore(pgStores.Tenants) in gateway startup
* feat(security): tenant-aware provider registry, event filter, and membership validation
- Refactor providers.Registry: Get(ctx, name) / List(ctx) extract tenant
from context via injected TenantFromCtx func (avoids circular import)
- Event filter: fail-closed 3-mode tenant filtering
Mode 1: unscoped admin sees all
Mode 2: scoped admin sees tenant events + system events
Mode 3: regular user sees only own tenant (fail-closed)
- WS connect: resolveTenantHint validates membership via GetUserRole
with PermissionCache (30s TTL, bus invalidation)
- BroadcastForTenant helper for tenant-scoped event emission
- Session list: add TenantID to SessionListOpts from context
- Cron handleRun: preserve tenant in background goroutine context
- GOCLAW_LOG_LEVEL env var (debug|info|warn|error) for Docker/K8s
- Cache debug logging: tenant_cache, permission_cache, api_key_cache
- Friendly verify error: timeout → user-readable message
- Verify timeout: 15s → 30s
* feat(ui): setup wizard improvements + agent preset enrichment
- Setup: skip link with confirm dialog, language selector (en/vi/zh)
- Setup: card padding fix (py-0 gap-0 on Card, py-5 on CardContent)
- Setup: remove duplicate skip link from layout
- Step Model: verify countdown timer (30s), stops on result
- Step Agent: default Fox Spirit preset, selected state styling,
hide agent key/name inputs, auto-derive from preset, emoji in config
- Summoning modal: elapsed timer (m:ss format)
- Agent presets: enriched prompts with human-like quirks
Fox Spirit: playful personality, care reminders
Artisan: portrait/banner/ads/logo expertise
Astrologer: reference sites (astro.com, cafeastrology, labyrinthos)
- i18n: "triệu hồi linh hồn" fix, all 3 locales updated
* feat(ui): API Key tenant support + card layout + provider chain fix
- API Key create: tenant selector for cross-tenant admin, provision scope
- API Key create: redesigned dialog with scope cards, Radix Select, icons
- API Key list: card layout with badges (status, tenant, scopes)
- API Key: shortcut in user menu (topbar)
- API Key: keep "API Key" untranslated across all locales
- Provider chain: empty state fix — skip legacy entry when provider
not found in current tenant
- i18n: form.cancel key added to all 3 locales
* fix(ui): add bottom padding to all page layouts + misc improvements
- Add pb-10 to all 24 page containers to prevent content touching
bottom edge of viewport
- Various UI polish from user modifications (summoning colors,
layout icon, agent cards, sidebar adjustments)
* feat(ui): MCP user credentials dialog + builtin tool tenant toggle
- MCP: per-user credentials dialog (api_key, headers, env KV editor)
with status badges, delete all, save
- MCP: "My Credentials" button on each server row
- Builtin Tools: per-tenant enable/disable override toggle
with "Using default" / "Enabled/Disabled for tenant" badges
and reset-to-default button
- Setup: larger logo (h-16) and bolder title (text-4xl font-bold)
- i18n: all keys added to en/vi/zh for both features
* fix(ui): API key card spacing + remove pagination border
- Card padding: px-4 py-3.5 (was px-3 py-2), rows spaced with gap-2
- Scopes on separate row from dates for readability
- Card gap: space-y-2.5 between cards
- Pagination: add className prop, remove border-t on API keys page
- Badge/icon sizes bumped to text-xs / h-3.5 (was text-[10px] / h-3)
* fix(security): comprehensive tenant isolation audit — SQL, events, cache, skills, files
Defense-in-depth hardening across 12 audit phases:
- SQL: add tenant_id WHERE to teams_tasks lifecycle/activity/followup/progress/embedding (~30 functions)
- Events: broadcastTeamEvent + task_ticker + subagent announce now carry TenantID
- Cache: agentKeyCache scoped by tenant (agent keys per-tenant, not globally unique)
- Skills: SkillStore interface accepts ctx, SQL filter (is_system OR tenant_id=$N), per-tenant list cache, GrantToAgent includes tenant_id, tenant-scoped file storage
- Files: StorageHandler/FilesHandler/TeamAttachments/teamWorkspaceDir use config.TenantDataDir/TenantTeamDir
- Security: HMAC signed file tokens (file_token.go) replace gateway token in URLs
- Audit: AuditEventPayload carries TenantID for async subscriber tenant scoping
- InboundMessage: subagent/dispatch/validation/session_send propagate TenantID
- Pending messages: DeleteStale scoped by tenant
* fix(security): skip gateway token in URLs with signed file tokens
toFileUrl() now skips appending ?token=GATEWAY_TOKEN when the URL
already contains ?ft= (HMAC signed file token). Prevents gateway
token exposure via browser history, logs, and referrer headers.
* fix(security): stop persisting auth tokens in session media URLs
mediaToMarkdown() now stores clean paths (/v1/files/path) without
any auth tokens. Previously embedded ?token=GATEWAY_TOKEN (or ?ft=)
into markdown which gets persisted in session messages DB.
Frontend toFileUrl() adds auth at render time — tokens never stored.
* fix(security): migration 027 strips leaked gateway tokens from session URLs
Adds cleanup step to tenant foundation migration: removes ?token=xxx
from persisted media URLs in session messages. Old code embedded the
gateway token; new code stores clean paths only.
* fix(security): sign file URLs at delivery time, not persist time
Add SignFileURLs() utility that finds /v1/files/ and /v1/media/ URLs
in content and appends HMAC signed ?ft= tokens before delivery.
Applied at 4 delivery points:
- WS agent events (OnEvent callback in gateway_managed.go)
- WS chat.history response
- WS sessions.preview response
- HTTP /v1/chat/completions response
Sessions store clean paths only. Tokens are generated per-delivery
with 1h TTL — never persisted in DB. Frontend toFileUrl() skips
appending gateway token when ?ft= is already present.
* fix: file token verify path must match signed path (/v1/files/ prefix)
SignFileURLs() signs the full URL path "/v1/files/{path}" but the
verify in files.go auth() was using "/{path}" (without prefix).
HMAC mismatch caused all signed file tokens to return 401.
* fix(security): scope storage size cache per-tenant
sizeCache was a single global entry — all tenants shared one cached
size. Changed to sync.Map keyed by tenantBaseDir so each tenant gets
its own cached size calculation.
* feat(ui): redesign API keys page — table layout + code snippet dialog
Replace card-based API keys list with table layout matching MCP Servers
pattern. Add "API Key Usage" dialog with tabbed code snippets (cURL,
TypeScript, Go) showing gateway connection examples with syntax
highlighting and copy-to-clipboard.
* fix(builtin-tools): seed media tools disabled, fix tenant toggle, add unconfigured warning
- Seed media tools with Enabled=false and no default provider settings
(user must configure provider chain before enabling)
- Fix provider chain form ghost entries: validate provider exists in
tenant before showing (parseInitialEntries new-format path)
- Fix double toggle: show only tenant override OR global toggle, not both
- Fix list API: merge tenant_enabled from builtin_tool_tenant_configs
into response when tenant-scoped (was always null)
- Add ListAll() to BuiltinToolTenantConfigStore for full override map
- Add amber warning banner for enabled media tools missing provider config
* feat(mcp): require_user_credentials setting + KeyValueEditor for user creds
- Add require_user_credentials setting in mcp_servers.settings JSONB
- Backend: skip MCP server in LoadForAgent when user lacks credentials
- Frontend: toggle in MCP form dialog, persisted in settings field
- Redesign MCP user credentials dialog: replace raw Textarea with
KeyValueEditor (sensitive key masking for auth/token/secret fields)
- Add settings to mcpServerAllowedFields for HTTP update
* fix(security): restrict cross-tenant to owner IDs, config to owners only
- Gateway token + non-owner user ID: admin role but tenant-scoped
(no cross-tenant access). Fallback: only "system" is owner when
GOCLAW_OWNER_IDS not configured (fail-closed).
- Config page (WS config.* methods): wrapped with requireCrossTenant
middleware — non-owner admins get permission denied
- Config sidebar link: hidden for non-cross-tenant users
- Logout: clear tenant_id and tenant_hint from localStorage
(prevents tenant scope leak to next user session)
- Refactor: LOCAL_STORAGE_KEYS.TENANT_ID/TENANT_HINT constants
* fix(ui): chat bubble contrast, login logo, tenant no-access UX
- Chat bubble: use --chat-bubble-user CSS var (darker orange, L=0.50/0.52)
with text-white for WCAG AA contrast (~5.5:1)
- Login page: logo h-20 w-20, title text-3xl font-bold
- Tenant selector no-access: shield icon + hint text explaining
user needs admin to add them to a tenant
- Sidebar: GoClaw text uses text-sidebar-primary (brand color)
* feat(contacts): merge/unmerge contacts to tenant users
Add API and UI for linking channel contacts to tenant_users identity,
enabling cross-channel user identification within a tenant.
Backend:
- POST /v1/contacts/merge — link contacts to existing or new tenant_user
- POST /v1/contacts/unmerge — remove merged_id from contacts
- GET /v1/contacts/merged/{id} — list contacts by tenant_user
- GET /v1/tenant-users — list users for current tenant
- Add display_name + metadata columns to tenant_users (migration 27)
- All endpoints enforce tenant isolation via context tenant_id
Frontend:
- Checkbox multi-select on contacts table
- Selection toolbar with Merge/Unmerge buttons
- Merge dialog: link to existing user or create new
- Link2 icon indicator for merged contacts
- i18n: en/vi/zh translations for merge section
* fix(security): add tenant_id to span and embedding_cache inserts
SpanData struct was missing TenantID field — all span inserts failed
with NOT NULL constraint violation after migration 027 dropped defaults.
Fix captures tenant_id from context at emit time (6 call sites in
loop_tracing.go + subagent_tracing.go), then includes it in both
CreateSpan() and BatchCreateSpans() SQL (25→26 columns).
Also fixes embedding_cache writeEmbeddingCache() which was missing
tenant_id in its batch INSERT — same class of bug.
Both use MasterTenantID fallback for backward compatibility.
* feat: Introduce tenant switcher UI and enhance multi-tenant architecture documentation.
* fix(security): enforce tenant scoping, fix session isolation and UI cleanup
- Force cross-tenant admins to always have a concrete tenant_id (default
MasterTenantID) instead of unscoped WithCrossTenant — prevents mismatch
between session listing (no filter) and writes (MasterTenantID fallback)
- Make agent router tenant-aware: Get(ctx, agentID) resolves agent for
the caller's tenant, preventing cross-tenant agent cache collisions
- Fix context.Background() in title goroutine and summarization — now
uses tenant-aware context (WithoutCancel) so titles and compaction
persist to the correct tenant
- Add read-only SessionStore.Get() method; replace GetOrCreate in auth
checks (preview/patch/delete/reset) to prevent phantom session creation
- Inject tenant from channel instance into inbound message processing
- Remove "All Tenants" option from tenant selector, topbar switcher,
and ws-provider auto-select — admin must always operate within a tenant
- Fix contacts page selection toolbar layout shift (always rendered)
- Widen MCP credentials sensitive header regex to catch API_KEY etc.
* fix(security): propagate tenant_id in consumer handlers and background ops
- InjectTeamDispatch: use context.WithoutCancel instead of context.Background
to preserve tenant_id while avoiding cancel propagation from HTTP/WS handlers
- handleTeammateMessage/handleSubagentAnnounce: inject tenant_id from msg
- Add nil guard for outcome.Result to prevent panic on agent-not-found
- Use BroadcastForTenant for EventTeamTaskFailed/Completed/LeaderProcessing
- Remove unnecessary WithCrossTenant in autoSetFollowup (ctx already scoped)
- resolveAgentByKey: accept ctx param for tenant-scoped agent lookup
- pending_messages: use request ctx instead of cross-tenant for GetDefault
* fix(security): tenant-scope EnsureContact and PendingHistory DB operations
- All channel EnsureContact calls now use tenant-scoped ctx instead of
context.Background (whatsapp, slack, discord, telegram, feishu, zalo)
- PendingHistory: add tenantID field, thread through constructors
- All PendingHistory DB ops (load, flush, compact, delete) use tenantCtx()
- Normalize timeouts: 10s for simple queries, 15s for batch writes
* feat(teams): auto-attach media, retry completed tasks, improve tool messages
- Auto-attach workspace media from any tool (create_image/audio/video) to
team tasks via loop-level hook, not just write_file interceptor
- Store absolute paths in team_task_attachments instead of relative
- Extend retry action to support completed tasks (reopen for follow-up)
- Context-aware comment result messages with next-action guidance for
leader vs member roles and task status
- All tool results include task_id for agent follow-up actions
- Use #N "subject" format instead of raw UUIDs in tool messages
* feat(multi-tenant): tenant isolation for media, events, providers and UI
- Tenant-scoped media store, event filter, provider registry
- Tenant header propagation in WS/HTTP clients
- UI: tenant-aware chat messages, markdown renderer improvements
- Protocol: tenant error codes and event definitions
* docs: add multi-tenant architecture documentation
* fix(teams): store absolute paths in team_task_attachments
- AutoAttachWorkspaceFile: use cleanPath consistently instead of raw absPath
- executeAttach: resolve relative paths to absolute via team workspace
- AfterWrite interceptor already uses filepath.Clean (verified)
* fix(teams): attachment download handles both absolute and relative paths
filepath.Join with an absolute att.Path discards the teamBase prefix,
causing path traversal check to fail and download to serve wrong file.
Now checks IsAbs first — uses path directly for new absolute entries,
falls back to legacy join for old relative entries.
* fix(teams): attachment download validates against workspace root not tenant dir
Absolute paths stored in DB don't match TenantTeamDir structure
(master tenant has no tenants/ prefix). Now validates absolute paths
against dataDir (workspace root) instead. Legacy relative paths still
resolve via TenantTeamDir as before. IDOR check on att.TeamID ensures
cross-team isolation.
* fix(teams): attachment download uses workspace root, not data dir
Files are stored under GOCLAW_WORKSPACE/teams/ but handler was passed
dataDir (GOCLAW_DATA_DIR) — completely different directory. Now passes
workspace. Legacy relative paths resolve via {workspace}/teams/{teamID}/{chatID}/{path}.
* fix(security): use HMAC-signed file tokens for attachment downloads
Replace gateway token exposure (?token=) with HMAC-signed short-lived
file tokens (?ft=) for team task attachment downloads — same mechanism
used by chat file URLs.
Backend:
- team_attachments auth: accept ?ft= signed token (priority 1), Bearer (priority 2)
- teams_tasks RPC: sign download_url with HMAC at delivery time
- Add fileTokenSecret to TeamsMethods, thread through wireChannelRPCMethods
Frontend:
- Use server-signed download_url from attachment data instead of ?token=
- Remove useAuthStore dependency from task-detail-dialog
* fix(security): decouple file token signing from gateway token
- Generate random 256-bit HMAC key at startup (crypto/rand, memory-only)
- All file signing/verification uses FileSigningKey() instead of gateway token
- Remove ?token= query param fallback from /v1/files/, /v1/media/, attachments
- Only ?ft= signed tokens and Bearer header accepted for file access
- Reduce file token TTL from 1h to 5min
- Frontend: remove gateway token from all file URLs and imports
- Note: tokens invalidate on restart (acceptable for 5min TTL + WS reconnect)
* fix(ui): use signed download_url for task attachments
- Add download_url to TeamTaskAttachment type
- Use a.download_url (server-signed ?ft=) instead of bare URL
* fix(security): tenant-scope team workspace paths + show user/tenant in topbar
- WorkspaceDir callers now use config.TenantWorkspace() to resolve
tenant-scoped base dir (non-master tenants get workspace/tenants/{slug}/)
- Fixes: all tenants previously wrote to global /app/workspace/teams/
without filesystem-level isolation
- Affected: loop.go (agent run), team_tasks_mutations.go (task creation)
- teams_workspace.go already correct (uses TenantTeamDir)
- UI topbar: show "userId (tenantName)" in user menu
* feat(ui): redesign task detail dialog with improved UX
- Split monolithic 343-line component into 5 focused files
- New header: subject as title, identifier + status badges above
- Metadata grid with soft bg-muted/30 background, priority icons
- Attachments: card-style with mime-type icons + proper download Button
- Description/Result: markdown rendering via MarkdownRenderer
- Comments: avatar circles + markdown rendering for content
- All sections collapsible with chevron + count badge
- Timeline: vertical dot-line pattern, collapsed by default
- Fix kanban card hover layout shift (opacity instead of display toggle)
* fix(security): tenant-scoped workspace paths and tool cache isolation
- Scope team workspace paths to tenant directory
- Add tenant isolation to tool cache and task reads
- Shell deny pattern improvements
- Agent resolver and context file tenant scoping
- Sidebar tenant/user display fix
- Add tests for workspace, boundary, and context file interceptor
* fix(ui): tenant visibility fallback, merge coming-soon, task detail tweaks
- Tenants page: try tenants.list (owner), fall back to tenants.mine
for regular users; hide create button for non-owners
- Merge contacts dialog: add coming-soon banner (i18n en/vi/zh),
disable form and submit button
- Task detail: collapse attachments/comments by default,
guard download_url before rendering
|
||
|
Luan VuandLuvu182
|
19d34aacec |
fix(mcp): add empty properties to object schemas for OpenAI compatibility (#321)
MCP tools that accept no parameters return {type: "object"} without a
properties field, which is valid per JSON Schema spec. However, OpenAI
API strictly requires "properties" on object schemas and rejects
requests with HTTP 400: "object schema missing properties".
Add default empty properties map when type is "object" and the MCP
server provides none.
Co-authored-by: Luvu182 <208665161+Luvu182@users.noreply.github.com>
|
||
|
Viet Tran
|
84650c5c14 |
feat(teams): attachments refactor, semantic search, improved prompting (#310)
* feat(teams): refactor attachments, remove team_message, add task comments UI Major team system refactoring: - Drop team_workspace_files, team_workspace_file_versions, team_workspace_comments, team_messages tables; replace team_task_attachments with path-based schema - Add denormalized comment_count/attachment_count on team_tasks for dashboard perf - Auto-track file writes as task attachments via WorkspaceInterceptor - Remove team_message tool entirely (tool, store, i18n, builtin_tools, MCP bridge) - Members communicate via task comments; approve/reject use comments for audit trail - Add commented/new_task notification types to TeamNotifyConfig - Enrich task completion announce with member comments - User-created tasks stay pending (backlog) — no auto-assign to leader - Configurable member request tasks (member_requests.enabled in team settings) - Structured task description template in TEAM.md for v2 leads - HTTP attachment download endpoint with IDOR + path traversal protection - Web UI: count badges on task list, comments section with input, download button - Team settings UI: completed/commented/new_task toggles, member requests section * feat(teams): priority dispatch, compact prompting, realtime comments - Priority dispatch: DispatchUnblockedTasks dispatches only 1 task per owner per round (highest priority first). Fixes cancel bug where CancelSession killed innocent queued tasks. - Prompt rework: Replace verbose Task Decomposition (25 lines) with compact Task Planning (8 lines). Add explicit UUID warning and sequencing guidance for weak models (Qwen, MiniMax). - Recent comments in dispatch: buildRecentCommentsSummary appends 3 most recent comments to re-dispatched tasks (reject, retry, stale). - Enrich comment event payload with TaskNumber, Subject, CommentText (truncated 500 runes, UTF-8 safe). - UI: Board subscribes to TEAM_TASK_COMMENTED for realtime comment_count badge updates. Task detail dialog auto-refreshes comments on event. - Tool description hint: guide models to write self-contained task descriptions with clear objectives and context. * perf(teams): add ListRecentTaskComments with SQL LIMIT Dispatch only needs 3 most recent comments — avoid fetching all. New ListRecentTaskComments(ctx, taskID, limit) uses ORDER BY DESC LIMIT N then reverses to chronological order. * feat(teams): add subject embedding for semantic task search + improve prompting - Add vector(1536) embedding column to team_tasks with HNSW index - Implement hybrid search: FTS (0.3) + cosine similarity (0.7) with graceful fallback - Auto-generate embeddings on task create/update, backfill existing tasks on startup - Wire embedding provider into PGTeamStore via gateway_setup - Change FTS from OR to AND with prefix matching for precise keyword search - Reduce search page size from 30 to 5 to save tokens - Rename migration 000023 → 000024, bump RequiredSchemaVersion to 24 - Update TEAM.md hints: prefer search over list, batch task creation with blocked_by - Add anti-pattern examples to prevent sequential task creation |
||
|
Luan VuandLuvu182
|
a7f5acc1e3 |
fix(security): harden SQL injection, MCP prompt injection, sandbox fallback, and file serving (#246)
- execMapUpdate: validate column names with strict regex to prevent SQL injection - HTTP update handlers: add field allowlists (agents, providers, custom_tools, mcp, channel_instances) - pqStringArray: properly escape array elements to prevent PostgreSQL array literal injection - scanStringArray: handle quoted elements in PostgreSQL array format - MCP bridge: wrap tool results as external/untrusted content to prevent prompt injection - File serving: block access to sensitive system directories (/etc, /proc, /sys, etc.) - Sandbox: fail closed when Docker unavailable instead of silent fallback to host - Shell deny: fix base64 --decode bypass, add host exec 1MB output limit - ILIKE queries: escape % and _ wildcards in knowledge_graph, custom_tools, channel_instances Co-authored-by: Luvu182 <208665161+Luvu182@users.noreply.github.com> |
||
|
viettranx
|
843b550651 |
feat: runtime packages UI, pkg-helper, configurable shell deny groups (#244)
Runtime package management with security hardening: - pkg-helper: root-privileged daemon for apk install/uninstall via Unix socket - HTTP API: /v1/packages (list/install/uninstall/runtimes), admin role required for writes - Shell deny groups: 15 configurable groups (per-agent overrides via context) - Packages UI: Web page for managing system/pip/npm packages with confirmation dialogs - Docker: privilege separation (root entrypoint → su-exec drop), init for zombie reaping - Security: umask socket creation, persist file validation, deny pattern hardening (Node.js fetch/http, Python from/import, curl localhost, sensitive env vars) - Auth: empty gateway token → admin role (dev/single-user mode) |
||
|
teexiiiandviettranx
|
99dd363b13 |
feat(mcp): lazy-activate deferred tools on direct call in search mode (#235)
* feat: Implement MCP manager for server connections, tool registration, and deferred tool loading for agents. * feat: Add tests for deferred tool activation logic within the tool registry and agent loop. * fix(mcp): prevent deny list bypass via lazy activation + fix idempotency race - Add PolicyEngine.IsDenied() to check deny patterns (incl. group: expansion) before allowing lazily-activated deferred tools to execute - Check IsDenied() in both single-tool and parallel execution paths in loop.go - Make ActivateToolIfDeferred idempotent by checking activatedTools before returning false, preventing concurrent goroutines from being blocked - Add tests for deny-on-first-call, group deny patterns, and idempotent concurrent activation --------- Co-authored-by: viettranx <viettranx@gmail.com> |
||
|
viettranx
|
3f2b6e258e |
chore(teams): remove deprecated delegation tools
Remove delegate_search, evaluate_loop, handoff from: - Seed data, system prompt, i18n keys/catalogs, channel events - Consumer handler (handleHandoffAnnounce), handoff route lookup - HandoffRouteData struct + PG implementation - Protocol events, MCP bridge comment - Web UI locale files (en/vi/zh) |
||
|
viettranx
|
27fb900510 |
refactor(tools): remove workspace_read/workspace_write, use file tools for team workspace
Remove dedicated workspace tools in favor of making existing file tools (read_file, write_file, list_files, edit) team-workspace-aware. - Delete workspace_tool_read.go and workspace_tool_write.go - Clean up workspace_dir.go: export WorkspaceDir, remove dead code (workspaceRelPath, sanitizeFilePath, inferMimeType, templates, etc.) - Remove workspace tool registration from gateway_managed.go - Remove workspace tool references from policy, subagent, MCP bridge - Add PathAllowable/PathDenyable to types.go for interface abstraction |
||
|
Viet TranandNam Nguyen Ngoc
|
1a42dc93a6 |
feat(teams): team system v2 with bug fixes, workspace scope, versioning, and prompt optimization (#183)
* feat(workspace): add team shared workspace for file collaboration
- Add workspace_write and workspace_read tools for agents to share files across team members
- Create team_workspaces DB table with migration 000017 (file metadata, pinning, tags)
- Implement PostgreSQL store layer for workspace CRUD operations
- Add RPC handlers for workspace list/read/delete from web UI
- Build React workspace tab with file listing, content preview, and delete
- Propagate workspace channel/chatID scope through delegation chain
- Auto-allow workspace tools in agent tool policy when agent belongs to a team
- Inject team workspace guidance into system prompt for team agents
- Add /reset command handler for clearing session history
- Harden MCP bridge context middleware to reject headers when no gateway token
- Add i18n strings for workspace UI in en/vi/zh locales
* feat(teams): add comprehensive task management with followup reminders and recovery
- Add task followup/reminder system with auto-set on lead agent reply and auto-clear when user responds on channel
- Add task recovery ticker to re-dispatch stale/pending tasks periodically
- Add task scopes, filtering by status/channel/chatID, and task events
- Add WS RPC handlers for task CRUD, assignments, comments, events, and bulk operations (teams_tasks.go)
- Add task detail dialog, settings UI for followup config, and scope filtering in web dashboard
- Add migrations 000018 (team_tasks_v2) and 000019 (task_followup)
- Extend team_tasks_tool with await_reply, clear_followup actions
- Auto-complete/fail team tasks when delegate agent finishes
- Add workspace file listing and team tool manager enhancements
* docs(teams): add team system architecture and playbook ideas documentation
- Add TEAM_SYSTEM.md with full architecture design covering task management, shared workspace, and delegation engine subsystems
- Add TEAM_PLAYBOOK_IDEAS.md outlining future team coordination layers (playbook, member capabilities, auto-learned patterns)
- Document data models, status flows, tool actions, followup reminder system, task ticker, execution locking, and workspace scope model
* fix(teams): resolve 6 critical bugs in team task system
- Fix unblock SQL: check array_length after array_remove (not before)
- Enforce single-team leadership in team creation
- Add requireLead() for approve/reject tool actions
- Validate cross-team dependency references in blocked_by
- Add team_id to handoff route for multi-team isolation
- Set blocked_by DEFAULT '{}' to prevent NULL array issues
* refactor(workspace): use stable userID as scope key instead of connection UUID
Workspace scope changed from (team_id, channel, chat_id) to (team_id, userID).
Fixes workspace fragmentation across WS tab refreshes and reconnections.
* feat(teams): add V1/V2 versioning with feature gating and optimized prompts
- IsTeamV2() helper gates advanced features (locking, followup, review, audit)
- V2 tool actions rejected for V1 teams with clear error message
- Ticker, gateway consumer, delegation hooks respect version flag
- TEAM.md renders v1/v2 sections conditionally
- Tool descriptions and params optimized (~38% token reduction)
- UI: version toggle in settings, V2 Beta badge, conditional rendering
- i18n: version modal keys for en/vi/zh
* fix(migration): use VARCHAR(255) for user ID columns and add metadata JSONB
- assignee_user_id, user_id, actor_id: TEXT → VARCHAR(255)
- Add metadata JSONB to team_task_comments and team_task_attachments
---------
Co-authored-by: Nam Nguyen Ngoc <namnn.0911@gmail.com>
|