Gemini drops tool_call_id and pairs functionCall/functionResponse by
function name, so its OpenAI-compat shim requires a non-empty
FunctionResponse.name. The name was only recoverable through a reverse
id->name lookup over assistant tool_calls still present in the request
window, which fails after pruning, truncation or tool_call collapse and
produces HTTP 400 "Name cannot be empty" on any follow-up iteration.
Add Message.ToolName, set it at every tool-result creation site and
preserve it through context pruning, then prefer it when serializing.
Fall back to the existing index for history persisted before the field,
so old sessions keep working without a reset.
When neither source resolves a name, drop the unlabelled tool result
instead of emitting an empty one: an empty name is a guaranteed 400 and
a synthetic name would match no prior functionCall. Gated behind the
existing Gemini detection so other OpenAI-compat hosts, which pair by
tool_call_id, are unaffected.
Field is JSON-optional, so session history needs no migration and older
binaries ignore it on rollback.
- Change default mode from "off" to "cache-ttl" (enabled by default)
- Increase softTrimRatio from 0.25 to 0.3 to match TS defaults
- Update tests to reflect new default-enabled behavior
Adds config.defaults WS method so UI can render pruning-knob placeholders
that mirror the server's internal fallback values. Exports
agent.DefaultPruningValues() as the single source of truth used by both
the resolver and this RPC — keeps UI in sync without a parallel table.
Port goclaw context pruning to match upstream TS design in
openclaw/src/agents/pi-hooks/context-pruning/:
- Opt-in default: prune only when mode="cache-ttl" (was opt-out)
- Remove Pass 0 per-result 30% guard (duplicated Pass 1 with different
suffix, caused wobble)
- Dedupe double prune call per iteration: PruneStage owns the single
entry point; loop_history only runs limitHistoryTurns + sanitizeHistory
- Add cache-TTL gate for Anthropic prompt cache: skip prune while cache
is live, scoped per-session via sync.Map
- Add context.pruned event emission for observability
- Configurable TTL as Go duration string ("5m", "30s")
BREAKING CHANGE: context pruning now opt-in. Add
contextPruning.mode: "cache-ttl" to config.agents.defaults to restore.
Migration 51 / SQLite v19 backfills mode="cache-ttl" for agents with
existing custom context_pruning config missing the mode field, so
previously-configured agents keep pruning after the opt-in flip.
NULL configs stay NULL (new opt-in default applies).
Web UI adds Cache TTL input + toggle wiring mode to cache-ttl/off.
- Replace char-based heuristic (chars/4) with tiktoken BPE for accurate
token counting, especially for non-ASCII content (Vietnamese/Chinese)
- Add pruningEstimator wrapper with tiktoken/fallback dual-path
- Raise default soft trim budget from 3K to 6K chars (3K head + 3K tail)
- Media tools (read_image, read_document, read_audio, read_video) get
higher soft trim budget (8K: 4K head + 4K tail) and skip hard clear
entirely — their vision/audio descriptions are irreplaceable
- Add per-result context guard (Pass 0): force-trim any single tool
result exceeding 30% of context window
Problem: Agent sessions accumulated 71K+ input tokens (83% history)
because read_file and exec had no output limits. SOUL personality
drowned by massive context.
Changes:
- read_file: add offset/limit params + 50K char output cap with
pagination hints (model can re-read with offset)
- exec/shell: cap output at 30K chars with smart head+tail truncation
(preserves errors/summaries at tail)
- pruning: add per-result 30% context guard, tune softTrimRatio
0.3→0.25 and softTrimMaxChars 4K→3K, add tail-aware soft trim
- mid-loop: allow pruning to re-trigger each iteration (was one-shot)
Design: cap at source, preserve full data in session, prune at
consumption time. read_file offset/limit enables recovery of
truncated content.
Root cause: compaction threshold compared total prompt tokens (including
system prompt, tool definitions, context files) against budget. Sessions
with only 50-60k of conversation could trigger compaction when overhead
pushed total to 120k+.
Changes:
- Enable context pruning by default for all providers (was Anthropic-only)
- Add EstimateHistoryTokens() that excludes system messages from count
- Fix EstimateTokens() to include ToolCall argument sizes
- Calibrate overhead on first LLM response, subtract from threshold checks
- Mid-loop: prune old tool results first (at 70% budget), compact only if
still over budget after pruning
- maybeSummarize: subtract overhead from calibration base
- Remove applyContextPruningDefaults() (Anthropic-only auto-enable)
- Simplify context pruning UI: show only Keep Last Assistants by default,
hide advanced params behind collapsible, remove Mode selector
Multi-agent AI gateway with WebSocket RPC, HTTP API, and messaging channel integrations.
Go port of OpenClaw with multi-tenant PostgreSQL, per-user isolation, security hardening,
and production observability.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Justin Truong