The fallback provider set was always catalog.all_provider_ids(), computed
at every call site and threaded through five layers alongside the catalog
itself. Fold it into Catalog::resolve_selection_with_catalog_fallback and
carry only a catalog_fallback flag through the transform/validate/
materialize entry points.
- materialize_run delegates to resolve_run_model again instead of
re-inlining its provider normalization and selection
- run_preflight derives ready providers from llm_result instead of
taking both, so callers cannot pass inconsistent pairs; the legacy
tests now exercise the production ready-first routing path
- AppState::resolve_llm_client_with_ready_ids replaces three copies of
resolve-then-extract-provider-ids, and ready_llm_provider_ids
delegates to it
- the unreachable "model resolution failed" preflight check becomes an
invariant error where the materialized run is produced
- validate_prepared_manifest_with_vars/_for_preflight share the
ValidateInput construction
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Replaces the per-field if-let cascade in the structured completion path
with a single struct-update expression. The cascade had to be extended
by hand for every request field and silently dropped stop_sequences and
provider_options, which the non-structured path already forwarded.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Unify list_events_from with list_events_from_with_limit so projection
replay shares the seek path instead of duplicating the decode loop
- Bound the event scan with keys::run_events_range instead of an
unbounded range plus a manual prefix break, so slatedb never touches
SSTs belonging to other runs or namespaces
- Store reader event_seq as None instead of a valid-looking sentinel of
1, so appends through a reader-built inner fail as ReadOnly rather
than writing duplicate sequences
- Borrow keys during scans instead of allocating a String per entry,
drop a dead branch in cached_events_from, collapse recover_next_seq's
single-caller parameters, and document the zero-padded key ordering
invariant the seek depends on
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The unknown-value rejection test duplicated strum coverage in
fabro-model and the HTTP 422 test in fabro-server. Keep only the
field-type assertion, using the same field-pinning idiom as
stage_model_usage_round_trip.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Adapted from poolside's official favicon mark: monochrome
fill="currentColor" at 24x24 to match the other provider logos, with the
brand's gradient-fade tail preserved via the original alpha mask.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Extract the quadruplicated watchdog check-and-clear logic in
schedule_worker_cancel_escalation into ManagedRun methods
(escalation_still_current, clear_escalation_for)
- Derive strum::IntoStaticStr for WorkerRef instead of a hand-written
variant-to-string match in kind()
- Use the generated AgentControlState constant instead of the raw
"waiting_for_steer" literal in run-detail.tsx
- Replace optimisticCancellationRunId state with a boolean; the
component is keyed by run id, so the stored id could only ever be
this run's own
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The crate reorganization renamed lib/crates/ to lib/apps|components|foundation/.
Git followed all modified files across the rename; the only conflict was the
newly added codec/cache.rs, now placed at lib/components/fabro-llm/src/codec/cache.rs.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Reuses run_multi_turn_cache_test — the same live cache verification the
anthropic, openai, and gemini routes already have. OpenRouter was the
one caching route with no live caller, which is exactly where the
missing-breakpoints bug hid: unit and wire tests prove we now send
cache_control, but only a live call proves OpenRouter forwards it to
Anthropic and cache reads actually appear.
Runs with: set -a && source .env && set +a && \
cargo nextest run -p fabro-llm --profile e2e --run-ignored only \
-E 'test(openrouter_claude_multi_turn_cache)'
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Anthropic prompt caching is opt-in per request: without explicit
ephemeral cache_control breakpoints in the body, no cache writes or
reads ever happen. The OpenAI-compatible codec never emitted them, so
every run on openrouter Claude models billed the full conversation at
the uncached input rate on every turn (0 cache tokens on the billing
page, confirmed by OpenRouter's activity portal).
- Add a `cache_control_breakpoints` model feature declaring that a
route only caches when the request marks the cacheable prefix; set it
on the builtin OpenRouter Claude rows. Catalog build rejects the flag
without `prompt_cache`.
- Teach the Chat Completions wire shape a parts-form content variant so
a message can carry the annotation; unmarked messages keep the
plain-string form for compatibility with strict servers.
- Mark the last system message (covers tools + system upstream) and the
second-to-last user turn, counting tool results as user turns —
mirroring the anthropic codec's placement so agent loops get
incremental cache hits.
- Extract the shared placement/opt-out policy into codec::cache and
refactor the anthropic codec onto it; anthropic wire snapshots are
unchanged.
- Honor `provider_options.<name>.auto_cache = false` as an opt-out and
consume the control key instead of merging it into the body.
- Mirror the new feature through settings (fabro-config), the OpenAPI
schema, and the generated TypeScript client.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
