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.
Rename the Artifact Generator display name to "Artifact" (the internal
artifact_generator key is unchanged) and tighten the user-facing one-line
descriptions for Code Executor, Artifact, and Read Document.
When code_executor or read_document is given an input that doesn't resolve to an
existing artifact, it now falls back to the caller's own chat attachments,
bridges the referenced one into a conversation-scoped artifact (idempotent,
server-computed size/sha256, quota-respected), and stages it. This is lazy —
plain chat uploads are never bridged, only a file a tool actually references —
and conversation-scoped only (workflow nodes bridge attachments up front).
The match is confined to the request's own attachments and re-verified against
the user-scoped attachments repo, so a model-supplied name/id can only ever reach
the caller's own files. Only the current request's attachments are reachable (the
attachments table has no conversation column); prior-turn files are a follow-up.
Give each produced artifact a short virtual handle - A1, A2, ... - the n-th
artifact in the conversation or workflow run (case-insensitive, not stored). The
tools return it in their results, and the edit, rewrite, and input parameters
accept either the handle or the full id. A handle resolves only within the
caller's own conversation or run, and the resolved id is still checked against
that parent before any read, so it cannot reach another tenant's artifact. This
fixes the model creating a duplicate instead of a new version when asked to
edit an artifact.
Add a code workflow node that runs code in the run-scoped sandbox session and
writes produced files as artifact references into workflow state, passing them
by reference (only id and metadata, never bytes) so downstream nodes and CEL
conditions can branch on them. Add an artifacts.* templating namespace that
resolves those references to metadata via a run-scoped lookup, available to
both the workflow engine and the prompt renderer. Extract the sandbox-to-
artifact persistence into a shared helper reused by the code node and the
code_executor tool.
Add a code_executor agent tool with a run_code action that runs agent-provided
code in the per-conversation sandbox session and captures produced files as
artifacts. Inputs are materialized only from artifacts the caller can access
(parent-scoped); produced files are stored under the user's namespace with
server-computed size and sha256, and the storage write is ordered last in the
transaction so a failure cannot orphan bytes. Output is a compact payload with
no raw bytes, and the produced artifact lights up the existing tool artifact
rail. Execution honors a wall-clock timeout and an agent-selectable session TTL
clamped by the global cap, and the action can be gated behind approval.
Tool-call argument logging is redacted so code bodies are not written to logs.
Alex