`CitationPolicyCapability` makes the single enforcement decision, whatever mix of
evidence capabilities is registered: two of them must not each demand a citation for
one answer. It decides in `after_model_request`, when a response carries no tool calls
and the question can still be redirected. An explicitly ungrounded answer is a
declaration and is left alone; a question that gathered no evidence is left alone too,
read from the ledger rather than from a searches dict that a new question clears. When
the cite tool is already withdrawn the question is recorded in
`CitationPolicyState.violations` instead of pointing the model at a tool that is gone.
Registering it is the only switch. `DiscoveredEvidence` and discovery move to
`capabilities.evidence` so both optional capabilities share them, and `cite_available`
joins `evidence_tool_names` as public for the same reason.
Measured on Qwen3.6-35B over two arms of 37 questions, 29 of them unanswerable from
the corpus: explicit ungrounded declarations rose from 23 to 26, grounded answers to
unanswerable questions fell from 4 to 2, answerable questions stayed at 8 of 8, and
one redirect fired in the whole arm. Reading every answer found no invented grounding.
The nemotron rows for retrieval and for both capabilities are re-measured on
this release with no reranker, judged by Qwen3.6-35B with thinking on. Rows for
other embedders and older versions keep their own attribution.
The capability pages said tool results from earlier turns are replaced before every
model request. That is now the compaction capability's job, and only when a host
registers it, so both pages point at it instead of describing it as automatic.
`RAGState` gains its `evidence` field, and the note about per-run resets now says
what a resumption keeps.
`_compact_old_tool_returns`, `PRIOR_TURN_NOTICE` and `turn_start` leave
`RAGCapabilityBase`, along with its `wrap_model_request` hook. The evidence
capabilities now retrieve and validate, and nothing else. Registering the compaction
capability is what rewrites a request; leaving it out sends the transcript untouched,
which was never a choice a host could make before.
The boundary is the recorded question identity rather than message shape, so a
resumption compacts what lies below the question in progress instead of switching
compaction off for the whole run. The newest earlier evidence return carries the
capsule and every other becomes a receipt, so one capsule exists by construction and
every return stays paired with its call.
Pictures of cited evidence are fetched through the capability that retrieved them and
re-attached beside the capsule with fresh labels. Ownership of a picture on the wire
requires the machine tag we write and an image directly after it, since neither
position nor prose is proof: several tools' results can arrive in one request, and a
user quoting our wording above their own picture had it removed. A picture that cannot
be fetched or decoded is emitted with neither its image nor its label.
The chat TUI and the example backend register the compactor, being multi-turn.
`client.ask`, `client.analyze` and the MCP tools do not: a single-shot question has
nothing earlier to compact.
The request-limit notice said only the cite tool remained available, but
chat registers rag and analysis in one agent, so exhausting analysis
claimed rag_search was gone too. Scoped to the capability's own tools.
The cite window was counted over every model request once loaded, so
turns spent on another capability expired it before the model was ever
placed where citing was the obvious move. Count only requests whose
preceding response called one of this capability's tools; engagement is
also the only thing that can loop, which is all the bound guards against.
Also: _count_tool_traffic returns a named tuple rather than four bare
ints, and counts failures only for this capability's tools, so host-tool
retries and output-validation retries no longer read as its failures.
- _budget_notice no longer names the cite tool after prepare_tools has
withdrawn it; the post-grace state gets the plain no-tools text back.
- Split search-budget rejections from any failed tool call: the code tool
raises ToolFailed for every error in model-written Python, so
budget_spent was true for a ZeroDivisionError.
- docs/capabilities/rag.md described the old single-turn removal.
- Drop the rationale clause from the CHANGELOG entry.
The pool watchdog counts only the time a worker spends running code, so a read
that blocks the worker never trips it. The margin above code_timeout therefore
guarded a race that cannot happen, and only bought a runaway call 90s where 60s
was configured. Pass code_timeout straight through. The two limits are disjoint:
the watchdog bounds a call that computes, and the read deadline bounds a call
that reads.
Drop the ordering test, which was true for any positive margin. The containment
test no longer sets code_timeout to zero, because that value now also disables
the watchdog and kills the worker before the read guard can refuse. It patches
the guard instead.
Monty 0.0.19 runs a plain class and a class with __enter__ and __exit__. Only
inheritance and metaclasses raise. Say that in the instructions.
MemoryFile is no longer constructed, so drop the import, the annotation, and the
stale references in the sandbox docstring and CLAUDE.md. That CLAUDE.md line
also still claimed a ThreadPoolExecutor, _run_async, and a fresh interpreter per
call.
The VFS bridge suspends the Monty worker for the length of a read. Monty checks
its duration budget between interpreter steps, so it cannot check while a read
is in flight. Code that reads in a loop overran a 60s budget by minutes. A read
takes about 20ms on a 2789-document corpus, so a full scan spends about 55s in
reads alone.
Check the deadline before each read. Raising from inside the callback answers the
worker's suspension, which keeps the session usable.
Monty also spends max_duration_secs across the session rather than per call, and
the sandbox reuses the session so that variables persist. Budget it for
code_timeout * max_executions. At the old per-call value the first slow call
starved every later one.
Do not wrap feed_run in asyncio.wait_for. Cancelling during pure compute is
clean, but cancelling while a read waits for an answer wedges the session with a
protocol RuntimeError that escapes execute(). A call that computes without
reading stays bounded by the session budget alone.
`docs/installation.md` claimed the full package contained every extra. It
pulls `docling`, `voyageai`, `cohere`, `zeroentropy`, `cross-encoder` and
`tui`, so `jina`, `s3` and `ingester` need installing separately. Without
`jina`, `provider: jina-local` falls back to no reranking. Anthropic is not
built in, and only the slim Docker image is published.
`docs/tuning.md` pointed at `claim_timeout_s`, which `WorkerConfig` now
rejects.
Drive analysis-only chat with analysis.model (falling back to qa.model) so
the running model matches the one the analysis capability configures,
including its vision flag; RAG-bearing chats still run on qa.model.
Give the AG-UI example state-bearing deps and a final STATE_SNAPSHOT so
registered citations reach the client, mirroring the app backend.
Update chat docs: haiku-rag chat uses --capability/-c, and drop the removed
"View state" command-palette entry.
reranking.multimodal (vllm provider only) attaches picture bytes to
synthetic picture chunks before rerank; VLLMReranker sends them as
content-parts documents (base64 data URI + description text) in the
same /v1/rerank request as plain text documents.
When a docling-serve instance crashes or returns 5xx, DoclingServeClient now
retries the request on a different instance (up to max_attempts) and trips a
per-instance circuit breaker so subsequent jobs skip a dead instance until its
cooldown elapses. Reuses the shared CircuitBreaker; adds max_attempts and a
nested circuit_breaker to providers.docling_serve.
Co-Authored-By: bryan davis <bryan@monkeytronics.org>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add `ingester.api.root_path` so the HTTP control plane (dashboard + API)
can be reverse-proxied behind a sub-path (e.g. /ingester/) on a shared
origin, instead of needing nginx sub_filter URL-rewriting.
- APIConfig.root_path: normalized ('', or single leading slash, no trailing
slash) via a field_validator; validate_assignment so CLI overrides
normalize the same way as config-file values.
- Forwarded to FastAPI(root_path=) and uvicorn.Config(root_path=) so
OpenAPI/docs links are prefix-aware.
- Dashboard route injects a <base href> matching root_path; all dashboard
fetches are now base-relative, so they resolve under the prefix while
staying identical at the root.
- `serve --root-path` CLI flag.
- Docs: "Behind a reverse proxy" section with an nginx example.
Closes#431
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Large files buffered entirely in RAM can OOM workers. Add
max_file_size to source config (default None = no limit).
FS checks stat().st_size before read_bytes(). HTTP and WebDAV issue
a HEAD request before GET when a limit is configured. S3 checks the
size from the existing head_async() call before get_async().
FileTooLargeError is classified as PermanentError so oversized files
go straight to the DLQ instead of retrying.
