fix: source authorization, silent retrieval failures, and prompt structure

Source access control
---------------------
`active_docs` is client-supplied and reached the retriever unchecked, and the
retriever queries `WHERE source_id = <id>` with no owner predicate — so any
caller could pass any source id to /stream or /api/answer and have another
tenant's documents quoted back, while /api/sources/<id>/search correctly
refused the same id. Gate it through `can_access`, the helper the guarded
endpoints already use, and filter `self.source` down to the authorized set.
Fails closed: no principal, or a check that errors, drops the source.

Three sibling paths had the same gap:

- workflow agent nodes: `AgentNodeConfig.sources` is written verbatim from
  client JSON at save time and nothing validated it, so a node could name any
  tenant's source. Gate against the workflow owner, so shared workflows keep
  reading their owner's sources like shared agents do.
- /api/share: `_resolve_source_pg_id` resolved any id with no ownership
  predicate and baked it into the agent the share creates; /api/search then
  searched it. Authorize before attaching.
- search_service: re-resolve the ids stored on an agent row instead of
  trusting them, so a row written by any future path with the same gap cannot
  be read back.

Team grantees previously lost their source's retrieval config: the post-check
read was still owner-scoped, so it missed and fell back to defaults (an
`agentic_tool` source was bulk-prefetched for every grantee). Read unscoped
after `can_access` passes.

Retrieval
---------
`PGVectorStore._ensure_table_exists` created an IVFFlat index on the empty
table it had just created. IVFFlat computes centroids at build time, so those
centroids were random, and combined with the `source_id` post-filter a source
with hundreds of embedded chunks returned zero rows — retrieval reported no
documents, the model answered from memory, and nothing was logged. Stop
creating the index (exact search is correct and fast well past the sizes most
deployments reach); raise `ivfflat.probes` to sqrt(lists) where an index still
exists; and re-run a short indexed search exactly, since post-filtering means
no index setting can guarantee a full result. `graphrag` had the same
empty-table index with no fallback at all.

Also: bound `chunks` to 0-500 on both the request and agent paths (0 still
means "skip retrieval"), let a source's configured `retrieval.chunks` outrank
the request body, and cap ClassicRAG's per-source floor at
max(top_k, n_sources) so attaching sources cannot inflate the result set.

Silent failures
---------------
An empty retrieval was invisible to both the model and the client: the `source`
event was suppressed when the list was empty, so "searched and found nothing"
looked identical to "no source attached", and the prompt said nothing at all.
Emit the event always, and tell the model when a search ran and returned
nothing. A file that parses to nothing now fails ingest with a message naming
the cause instead of storing an embedding of the empty string. `score_threshold`
returns warnings when the active store or retriever cannot honour it.

Prompt structure
----------------
Retrieved documents move from the system prompt into the user turn, with the
injection guard restated next to them: they change every turn (defeating prefix
caching), they are third-party text that should not carry system authority, and
routing them through the query budget makes them truncatable rather than
silently crowding it out. Documents are shed lowest-ranked-first before the
question is touched.

The six chat presets (3 tones x 2 retrieval modes) differed only in their
Answering section; they are now composed from single-source fragments at load
time, not through Jinja inheritance, which would have opened a file-read
surface in the template sandbox and broken the tool-prefetch parser. Per-tool
guidance moves out of the prompt into tool schemas, so it travels with the tool
and cannot render when the tool is absent. A plain-text custom prompt is staged
as a persona value inside the skeleton instead of replacing it wholesale — it
used to silently lose the injection guard, platform block, memory and
attachments, and its braces are now inert.

Other fixes
-----------
- agents/base: an oversized system prompt drove the query budget negative and
  dispatched a full-price request with an empty question; raise instead.
- llm/anthropic: migrate off the retired Text Completions API. It flattened
  history to first+last message and ignored tools entirely. Adds the missing
  Anthropic handler, without which every tool call was silently dropped.
- sources/upload: `sitemap` had no branch, so every sitemap ingest died on a
  TypeError; `validate_url` now rejects a falsy URL cleanly.
- workflow nodes: retrieved documents never reached the node agent, so a
  classic node with a source and an ordinary prompt answered "I have no
  documents" while the run reported completed.
- parser/bulk: copy the metadata dict, or every chunk reports the last chunk's
  token_count.
- crawler_loader: carry the page title, or citations render the whole chunk
  body as the label.

tests/api/answer/services/test_stream_processor_utils.py

@@ -384,14 +384,61 @@ class TestConfigureSource:
         sp._configure_source()
         assert sp.source == {}
 
-    def test_request_active_docs_used(self):
+    def test_request_active_docs_used(self, monkeypatch):
+        import application.api.answer.services.stream_processor as sp_mod
         from application.api.answer.services.stream_processor import (
             StreamProcessor,
         )
+
+        monkeypatch.setattr(sp_mod, "can_access", lambda *a, **k: True)
         sp = StreamProcessor({"active_docs": "abc"}, {"sub": "u"})
         sp._configure_source()
         assert sp.source == {"active_docs": "abc"}
 
+    def test_request_active_docs_the_caller_cannot_read_is_dropped(self, monkeypatch):
+        """``active_docs`` is client input; the retriever has no owner predicate.
+
+        An unchecked id read another tenant's documents straight into the
+        answer, while /api/sources/<id>/search correctly refused the same id.
+        """
+        import application.api.answer.services.stream_processor as sp_mod
+        from application.api.answer.services.stream_processor import (
+            StreamProcessor,
+        )
+
+        monkeypatch.setattr(sp_mod, "can_access", lambda *a, **k: False)
+        sp = StreamProcessor({"active_docs": "someone-elses-id"}, {"sub": "u"})
+        sp._configure_source()
+        assert sp.source == {}
+        assert sp.all_sources == []
+
+    def test_mixed_access_keeps_only_the_readable_ids(self, monkeypatch):
+        import application.api.answer.services.stream_processor as sp_mod
+        from application.api.answer.services.stream_processor import (
+            StreamProcessor,
+        )
+
+        monkeypatch.setattr(
+            sp_mod, "can_access", lambda conn, kind, sid, user: sid == "mine"
+        )
+        sp = StreamProcessor({"active_docs": ["mine", "theirs"]}, {"sub": "u"})
+        sp._configure_source()
+        assert sp.source == {"active_docs": ["mine"]}
+
+    def test_access_check_failure_fails_closed(self, monkeypatch):
+        import application.api.answer.services.stream_processor as sp_mod
+        from application.api.answer.services.stream_processor import (
+            StreamProcessor,
+        )
+
+        def _boom(*a, **k):
+            raise RuntimeError("db down")
+
+        monkeypatch.setattr(sp_mod, "can_access", _boom)
+        sp = StreamProcessor({"active_docs": "abc"}, {"sub": "u"})
+        sp._configure_source()
+        assert sp.source == {}
+
     def test_request_active_docs_default(self):
         from application.api.answer.services.stream_processor import (
             StreamProcessor,
@@ -509,7 +556,31 @@ class TestGetPromptContent:
 
         with _patch_db(pg_conn):
             content = sp._get_prompt_content()
-        assert content == "My prompt content"
+        # A plain-text custom prompt is staged as a persona value inside the
+        # composed skeleton so it keeps Boundaries/platform/memory rather than
+        # replacing them wholesale.
+        assert sp._persona == "My prompt content"
+        assert "## Boundaries" in content
+        assert "{{ system.persona }}" in content
+
+    def test_templated_custom_prompt_is_left_alone(self, pg_conn):
+        from application.api.answer.services.stream_processor import (
+            StreamProcessor,
+        )
+        from application.storage.db.repositories.prompts import (
+            PromptsRepository,
+        )
+
+        prompt = PromptsRepository(pg_conn).create(
+            "u", "p2", "Answer as of {{ system.date }}.",
+        )
+        sp = StreamProcessor({}, {"sub": "u"})
+        sp.agent_config = {"prompt_id": str(prompt["id"])}
+
+        with _patch_db(pg_conn):
+            content = sp._get_prompt_content()
+        assert content == "Answer as of {{ system.date }}."
+        assert sp._persona is None
 
     def test_returns_none_on_missing(self, pg_conn):
         from application.api.answer.services.stream_processor import (
@@ -559,7 +630,9 @@ class TestGetPromptContent:
         sp = StreamProcessor({}, {"sub": "u"})
         sp.agent_config = {"prompt_id": "default", "agent_type": "classic"}
         content = sp._get_prompt_content()
-        assert "source.summaries" in content
+        # Classic presets ground on documents supplied with the question;
+        # agentic ones instruct the model to call the search tool.
+        assert "documents are provided with the question" in content
 
     def test_null_prompt_id_agentic_agent_gets_agentic_preset(self):
         from application.api.answer.services.stream_processor import (
@@ -582,7 +655,9 @@ class TestGetPromptContent:
         sp.agent_config = {"prompt_id": None}
         content = sp._get_prompt_content()
         assert content is not None
-        assert "source.summaries" in content
+        # Classic presets ground on documents supplied with the question;
+        # agentic ones instruct the model to call the search tool.
+        assert "documents are provided with the question" in content
 
 
 class TestPreFetchDocs: