fabro-llm's catalog module held some 250 lines of listing and picking
helpers over lithos data: enabled and listed providers, model lookup by
id, alias, or wire id, matches ranked as the resolver ranks, default and
probe models, the small utility model across ready providers, the nearest
model on another provider, and cost by handle. lithos-llm now answers all
of those on `Catalog` and `CatalogProvider` through `Offering`, so the
helpers and the `ModelEntry` wrapper go.
What stays in Fabro's catalog module is its own: building the catalog from
the operator overlay, and reading the agent harness and
`reasoning_by_default` from the shared `metadata.agent` namespace. The
passthrough selection policy in `selection.rs` keeps its rules and calls
lithos for the lookups.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
fabro-auth defined its own `CredentialSource` trait beside the lithos
`CredentialProvider`, with a parallel `ResolveError` and an adapter between
them, because lithos had no way to ask which providers a store can serve
right now. It does now: `credentials::readiness`, `ClientBuilder::build_ready`,
and `CredentialError::Unusable`.
- The vault, SQL vault, API-key, and extra-headers stores implement
`CredentialProvider` directly. Material that is present but unusable (an
expired token with no refresh, a wrong-typed vault entry, a header secret
that did not resolve, a store read failure) is `CredentialError::Unusable`
with the operator-facing reason; its `Display` replaces
`auth_issue_message`. `is_configured` is the cheap presence check.
- `fabro_llm::build_client` calls `build_ready`; `FabroClient::auth_issues`
carries `CredentialError`. `fabro_llm::configured_providers` replaces the
per-store `configured_providers` method.
- `CredentialSource`, `ResolvedCredentials`, `lithos_credentials`,
`ResolveError`, and `auth_issue_message` are deleted. Twenty files that
held `Arc<dyn CredentialSource>` hold `Arc<dyn CredentialProvider>`.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
lithos-llm now exposes `ReasoningEffort::ALL`, `Speed::ALL`, `as_str`,
`Display`, and `FromStr` on its request-control enums. Fabro's
`controls` module kept parallel name tables and parsers for them; only
the nearest-supported-effort rule is Fabro's own, so that is what stays.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Fabro's policy layer restated the lithos built-ins under `metadata.fabro`:
enabled flags, credentials, display facts, probe and small-default roles,
and agent profiles. lithos-llm now carries every one of those as a core
field or under the shared `metadata.agent` namespace, so the layer and its
typed view go:
- Delete `fabro-policy.toml` and `FABRO_POLICY_TOML`. The catalog is the
lithos built-ins plus the operator's `[llm]` overlay, nothing between.
- Delete `fabro_types::catalog_policy`. `enabled`, `stands_in_for`,
`api_key_url`, `family`, the cutoffs, `estimated_output_tps`,
`small_default`, and `probe` are read from lithos accessors; the agent
profile and `reasoning_by_default` come from `metadata.agent`, which
Pebble reads too.
- `catalog::provider`, `enabled_providers`, and `listed_providers` return
the lithos `CatalogProvider` directly; `ModelEntry` loses its policy
field and gains `agent_profile()`.
- Test fixtures move `[providers.x.metadata.fabro] enabled = true` onto
the provider table, drop `credentials` lists in favor of the secret name
lithos derives from the provider id, and spell `agent_profile` as
`metadata.agent.profile`.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
lithos-llm now owns which named secrets each provider reads and how they
shape into its auth scheme, including a derived `<PROVIDER>_API_KEY` for
operator-defined providers. Fabro's job shrinks to supplying the store:
`VaultCredentialSource` hands lithos a lookup that reads the process
environment, then the vault, under the same conventional names.
What Fabro still adds on top: the Codex OAuth credential in the vault,
refreshed and persisted when it expires; `{{ secrets.NAME }}` tokens in a
provider's `default_headers`, resolved against the vault and re-sent as
credential headers; and OpenAI organization and project headers from the
environment.
Deleted with the `metadata.fabro.credentials` list: `CredentialRef`,
`CredentialResolver`, `EnvCredentialSource` (now
`VaultCredentialSource::environment_only`), and the `env_var_names` /
`expected_vault_secret_name` helpers, replaced by `secret_names` and
`expected_secret_name` over the lithos table. `openai-codex` joins the
first-party provider id constants.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
lithos-llm sends the application name as the codex `originator` header,
so the OpenAI Codex deployment can tell which harness a request came from.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
lithos-llm now owns `enabled` and `stands_in_for` as core provider
fields, and its `CatalogResolver` refuses disabled providers and reroutes
a request to the provider standing in for an unavailable one. Fabro's
resolver re-implemented both from `metadata.fabro`, so it goes.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
lithos ships `metadata.agent.profile = "gpt6"` on the GPT-6 Astra row.
Fabro runs it on the GPT-5.6 harness: the same Codex core tool set,
memory filenames, question tool, and command timeout. The shared
`uses_codex_core_tools` predicate replaces the `== Gpt56` checks so the
two kinds cannot drift apart.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Workflow LLM handlers build lithos requests, bill from lithos usage and
cost, and classify failures from lithos `ErrorKind`. Model resolution and
fallback use the fabro-llm selection and catalog helpers. Validation
rules read the lithos catalog, and store fixtures use the new
`BilledModelUsage` shape.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
fabro-llm is now a thin integration crate: catalog construction from the
lithos built-ins, the Fabro policy layer, and operator overlays; client
construction from catalog plus credentials; a Fabro `ModelResolver` that
enforces `metadata.fabro` policy; model selection; a server gateway
adapter; attachment inlining middleware; reasoning normalization;
one-shot structured output; probe wiring; and catalog API views. The
in-house codecs, transports, providers, tool loop, retry, cost, and
token-count code are deleted along with the wire snapshots that covered
them.
fabro-agent consumes lithos `StreamEvent`s and `Response`s directly.
Retry is split: lithos's retry middleware handles failures before any
visible output, and the agent replays the turn after.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Add lithos-llm as a pinned git dependency and replace the in-repo
`test/twin/openai` crate with the published `twins` revision that
lithos-llm verifies its codecs against. Move the Fabro policy overlay
(`fabro-policy.toml`) into fabro-llm so Fabro owns its own catalog
policy layer.
Drop the twin-openai nextest overrides and CI package filter now that the
crate is no longer a workspace member.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Move the "which failures can happen before Starting" classification onto
FailureReason as an exhaustive predicate and use it for every
Runnable -> Failed transition, replacing the hand-maintained allowlist.
Give the pending-cancel precedence rule a single owner shared by the
worker launch and worker exit paths.
Test cleanups: share the Notify wait loop, server record fixture, and
post-failure assertions; simplify the pre-start test runtime's hold
flag; and parameterize the slate run.failed payload helper.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Move ProjectedRun next to EventProjectionCache in run_state, since the
summary store both produces and consumes it, and give it a replay
constructor that owns the events-to-head derivation. load_projection now
returns RunNotFound directly instead of erasing it to None and having
callers rebuild it; load_run_projection is the single Option translation
point. install_in_memory_state reuses the existing From impl, and the
commit path passes its Arc through instead of unwrapping and
reallocating it.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The supervisor's periodic recovery scan asked SQLite for every run that
had ever recorded an explicit pull request creation request and then
replayed each inactive candidate's full history to learn whether the
request was still pending. With projections now loaded on demand that
set grows without bound and was replayed every scan.
The candidate query now mirrors the projection reducer: a run is a
candidate only when its latest creation request has no later request,
created, linked, or unlinked event, and no later failure naming the same
creation id. Callers still replay each candidate to confirm, so the query
only has to avoid omitting a pending run, and the replayed set is bounded
by in-flight requests.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Drop the unreachable active-run mismatch guard that was copied into
load_run_projection: the active-runs map is only ever inserted under the
handle's own run ID, so the check could never fire. Remove it from the two
pre-existing sites too and delete matches_run.
Trim install_in_memory_state to take only the committed projection, since
the event envelope duplicated last_seq and the inner scope only existed to
release the lock before the now-removed shared cache update. Add a From
impl so RunDatabase::build no longer hand-builds EventProjectionCache, and
rename projected_state_locked to match its projection_snapshot sibling.
In fabro-server, have reject_if_archived and ensure_run_exists read the
run summary row instead of replaying the full event history for inactive
runs; the summary is written in the same transaction as the event.
Fold the repeated store-reopen fixtures in fabro-store and fabro-server
tests into helpers, and fix a stale comment about the deleted shared
projection cache.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- Load recovery candidates from the warm projection cache instead of
replaying each run's full event history, and check dispatch
eligibility before any I/O
- Extract the shared can_dispatch predicate used by both the recovery
scan and the worker dispatch loop
- Drop load_durable_run_status, now identical to durable_run_status
- Share parse_stored_run_id across the three stored-id decode sites
- Reuse push_order for the canonical run ordering in list_all and
list_by_statuses
- Replace the test-only queue clear accessor with the existing drain
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Drop the post-decode run_id/session_id/event-body checks in
find_session_owner: decode_event_row already verifies every stored
column against the decoded envelope, and the WHERE clause pins
session_id and event_name to the requested values.
- Build the lookup query from SELECT_EVENT_COLUMNS like the sibling
event queries instead of duplicating the column list.
- Carry the stored run_id text in the unparseable-id error instead of
an "<invalid>" placeholder.
- Fetch applied migration versions once per migrate() and share the
set between the session-owner preflight and the pre-migration
snapshot; check the applied version first so steady-state startups
skip the sqlite_master probe. Mark the preflight as removable with
the run-history compatibility window.
- Restore session_by_id_key as a #[cfg(test)] helper so tests stop
hand-rolling the legacy reverse-index key shape.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Share canonicalize_location and resolve_existing_workflow_location
between the local package resolver and the version collector, drop the
redundant package-root pre-check and the PackageFileReadError enum in
favor of anyhow context, and read HEAD's SHA from git2 instead of a
separate rev-parse subprocess.
Make GitRunTargetObservation a plain struct, replace the repo-info
tuple with a named struct, tighten the closure view trait, avoid
deep-cloning the root workflow during collected validation, remove the
unused into_closure accessor, and dedupe test helpers.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Collapse the role-paired materializer error variants into
`Credentials`/`Checkout` tagged with a `CheckoutRole`, route both
checkouts through one resolve-then-prepare helper, and replace the
test-only clone-URL field on the production materializer with a
`GitRemote` resolver seam. A workflow source in the target's repository
now reuses the already-resolved credentials instead of minting a second
token.
Also inline the one-line workflow-source normalizer, drop the `as_str`
wrapper on the new kind enum, remove the unused migration constant, move
rather than clone scheduler fields, and deduplicate the web form's
ref-validity rule and per-kind copy into a single table.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Share the clone-based provider predicate and provider label between the
automation form and environment settings instead of duplicating them
- Hoist repeated environments query state in the new-automation route
- Normalize empty environment ids to None so validation needs one check
- Merge the scheduler's record/clear error helpers and skip the clearing
write when no error is stored
- Guard the environment backfill with a cheap existence query
- Drop an unneeded id clone and a no-op migrator comment
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Automations now store an environment_id that must reference an enabled
Docker or Daytona environment. Each trigger fire resolves the current
environment definition and snapshots its settings into the run, and
deleting an environment still referenced by an automation is rejected
with a conflict.
Existing automations are backfilled conservatively: a compatible
environment named default is selected when present, otherwise the sole
compatible environment. Anything ambiguous is left incomplete and cannot
run until an operator selects an environment in the web UI.
Scheduler failures are recorded on the automation as last_error and
cleared after the next successful scheduled run.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Hoist the SQLite backup/integrity helpers duplicated between the blob and
run-history activation migrations into one shared module, drop the
write-only recent-events buffer and other dead state left behind by the
SQLite cutover, reuse existing helpers for projection bootstrap, sequence
allocation, catalog keys, and test pool construction, and collapse the
tombstone and activation-marker plumbing to single statements.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
`DaytonaSnapshotSettings` carried two independent `Option`s (`image` and
`dockerfile`) that every consumer had to re-validate. Replace them with a
single `source: DaytonaSnapshotSource { Image, Dockerfile }` so the
both-set and neither-set states are unrepresentable at the sandbox layer.
This removes four unreachable error arms in `canonical_manifest` and
`create_snapshot_params`, the `.filter(...)` guard in `initialize`, and
the presence guard in `daytona_config_from_environment`. The
mutual-exclusion rule now lives only in fabro-config, which owns the
`image.docker` / `image.dockerfile` keys the old messages named.
Merge `ImageSnapshotManifest` into `SnapshotManifest` via a flattened
`SourceManifest` enum. The dockerfile case serializes to the same bytes
as before, so existing snapshot names are unchanged; the pinned identity
test still passes. Pin the image-case identity as well so a future
manifest change cannot silently orphan image snapshots.
Fold `validate_daytona_image_settings` into the existing Daytona arm of
`validate_provider_capabilities`; both callers already invoke it right
after `resolve_environment_fields`, so error order is unchanged.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
