Commit Graph
867 Commits
Author SHA1 Message Date
sam e7afe2da8b feat(eda): playbook document + event coordinator, single-action workflow-kind dispatch
Seventh slice of docs/prds/eda-playbook-engine.md (t2db.15), consuming
t2db.14's expr package: the rich trigger+actions playbook document and
the coordinator that dispatches it.

- internal/domain/eda/playbook: Load(dir) reads *.yaml/*.yml playbook
  documents (trigger + exactly one workflow-kind action + optional CEL
  when), validating the HARD BOUNDARY at load: a playbook with 2+ actions
  or a non-workflow action position is a reported load failure --
  multi-action and Module/Channel/Forge actions stay blocked on the
  unresolved execution-time gaps. Any malformed YAML, invalid trigger,
  missing workflow name, or when compile failure fails the WHOLE load
  (reject-at-load, same shape as routing collisions); a missing dir is an
  empty store. Trigger reuses the existing workintake kind/label
  vocabulary (no second label-matching mechanism). Dispatch matches
  trigger, evaluates when (CEL, event context with actions: {} -- no
  prior actions in single-action playbooks), and returns the named
  workflow from the registry via the same lookup Route() uses -- no
  duplicated dispatch logic. when eval error follows the resolved doc's
  J3: false, skip, caller logs.
- Store.Dispatch(reg, DispatchInput) is the coordinator entrypoint; input
  carries Labels/Kind/Event. Production task intake wiring
  (daemon.go/pollNATS/publishTask) is explicitly NOT in this ticket -- a
  follow-up; proven end to end via direct tests plus the shipped example
  (examples/eda-playbooks/bug-tdd.yaml) loaded and dispatched through the
  real path.
- Config: EDAPlaybookDir (eda_playbook_dir), documented as DISTINCT from
  PlaybookDirs (flat binding files) -- a reader cannot confuse the two.
- Bootstrap: playbook.Load wired in loadWorkflows, log-and-abort on
  failure matching every prior slice; nonexistent dir is an empty store.
- workflow.Route(), workflowForLabels, and the t2db.9-.12 binding loaders
  are UNTOUCHED (verified: their full test suites pass, zero regressions).

Also carries small lint cleanups on t2db.13/.14 code flagged by the gate
run: expr Eval no longer type-asserts to ref.Val (Eval returns ref.Val
directly in v0.32.0; cel-go's own Eval wraps evaluation in recover --
verified in source); module decodeLogArgs extracted to clear the nestif
finding. golangci-lint: 0 issues.

Tests (10): single-action round trip, when-false skip, 2-action load
failure, non-workflow position load failure, when compile failure drops
playbook, empty trigger load failure, trigger mismatch skip, no-when
match, one-invalid-file fails whole load, missing dir empty store, plus
the shipped example load+dispatch. task check clean (113/113 UI, exit 0).
2026-09-03 04:05:59 +10:00
sam 7e34fcb598 feat(eda): CEL expression environment (compile/validate at load, evaluate at dispatch)
Sixth slice of docs/prds/eda-playbook-engine.md (t2db.14): the
internal/domain/eda/expr package implementing the resolved open question 1
(CEL decision), standalone -- no coordinator/dispatch yet, mirroring how
t2db.13 built ModuleRegistry before any consumer.

- NewEnv: CEL env declaring the two context roots (event + actions, both
  map(string, dyn)) with DefaultCostLimit (100,000, tunable per J5).
- Compile: parses/type-checks an expression string; syntax error, unknown
  root (anything other than event/actions), or type error is a returned
  error -- the reject-at-load entry point the playbook loader and lint
  tool will share.
- Eval: evaluates a compiled program against a dispatch-time Context;
  missing-field/wrong-type/cost-limit errors are returned, never panics.
  No recover() wrapper: CEL is panic-free by design, verified by the
  hostile-data tests (5000-deep map, no panic). CEL's ref.Val wrappers
  (types.Bool etc.) are unwrapped to native Go values.

Dependency: cel.dev/cel-go v0.32.0 (direct), the only new dependency,
already vetted in the resolved doc's vetting record -- no re-vet.

Tests (table-driven, verified against real CEL semantics, not guessed):
valid boolean against event fields; syntax error at compile; unknown root
rejected at compile naming the root; missing event field errors not
panics (== on a missing key evaluates false -- verified CEL semantics);
has() presence macro; actions context reads; deep-nested hostile data
error-free or false without panic; cost-limit enforcement proven by
aborting a macro expression under a tiny limit (CEL's cost model accrues
on map/filter iteration, not field reads -- verified by probing).

task check clean (113/113 UI, exit 0).
2026-09-03 04:01:21 +10:00
sam cba3a9a86d feat(eda): Module position first slice -- registry + side-effect-free log kind
Fifth slice of docs/prds/eda-playbook-engine.md (t2db.13), implementing
docs/prds/module-position.md's recommended first slice: prove schema ->
go:generate -> Yaegi load -> ModuleRegistry end to end with a
side-effect-free action kind, before any kind with a real external effect
(notify, image-gen, forge writes) -- those stay blocked on the parent
doc's unresolved execution-time gaps (idempotency, mid-run failure).

- internal/domain/eda/module/log: hand-written schema source (Args/Result/
  SchemaVersion per module-position.md) + logextract generated symbol
  table via go:generate yaegi extract (same tooling as wfextract/
  gateextract; scoped to ./internal/domain/eda/module/...).
- internal/domain/eda/module: ModuleRegistry with Register(kind, dir) /
  Invoke(ctx, kind, rawArgs). No generic Module interface with an any
  payload: each kind is its own tiny package with its own generated
  contract; the registry's internal storage is the only place erasure
  appears (exactly as inside wfextract). Invoke decodes rawArgs strictly
  into the kind's Args before calling the interpreted function -- an
  unknown key or wrong type is a reported failure, not a silent
  zero-value fill (the 'schema defines the accepted message' rule).
  Generalizes gateeval's fixed-signature Yaegi pattern via
  yaegiutil.New/Resolve/Safe -- no new interpreter-construction or
  panic-recovery mechanism.
- Config: ModuleDir mirroring PluginDir's shape and doc comment
  (operator-trusted directory of Yaegi module implementations, one .go
  per kind; empty means none loaded). Wired at daemon startup in
  bootstrap.go alongside PlaybookDirs loading -- log-and-abort on load
  failure (a broken module is a startup failure, not degrade-and-skip).
- examples/modules/log.go: operator-facing example (//go:build ignore,
  mirroring examples/secret-engines/) proving the shipped contract.

Tests: registry (register/invoke, unknown kind, missing file, strict
arg-decode shape failures, malformed file, panic recovery), logextract
smoke test, and an example-load test that runs the shipped
examples/modules/log.go through the real registry (not a test fixture).
task check clean (113/113 UI, exit 0).

Playbook YAML action dispatch is deliberately out of this slice: a
minimal 'run this one module action, no chaining' dispatch still requires
naming how args values reach the action (parent doc open question 1 --
data-flow/interpolation syntax) and what the result feeds, both
unresolved by design. Dispatch is a follow-up ticket that consumes this
registry as-is; recorded in module-position.md.
2026-09-03 03:35:18 +10:00
sam 43c6f257ed fix(extract): regenerate stale Yaegi symbol tables; fix wfextract directive
Two pre-existing issues discovered while scoping the Module-position slice
(archie-core-e7bo), unrelated to the playbook engine work:

1. wfextract's //go:generate directive still targeted
   internal/workflow, a path gone since the domain migration -- the
   committed generated file was hand-patched for imports only, never
   regenerated. 'go generate ./...' failed repo-wide, silently.

2. Regeneration surfaced real drift on three surfaces:
   - wfextract: the committed table predated the migration and was
     missing everything added since -- routing YAML loaders (t2db.9/.10/
     .11), the review types, and the Forger/Reviewer/Trees interfaces
     that TaskContext exposes to every interpreted stage plugin. A stage
     plugin referencing tc.Trees or the review types could not resolve
     them.
   - pluginextract: internal/plugin/host.go (f73b2d1) added the
     Host/Module/Manifest/Permission/Health/CapabilityKind/LifecycleState
     /ModuleStatus types and NewHost/AdaptLegacy/HostAPIVersion, none of
     which were ever extracted. A daemon plugin written against the
     capability-host API (bootstrap.go:707 LoadDir) silently had no way
     to reference them.
   - secretextract: symbol exports were current; only the hand-applied
     nil-guards and comments differ from raw generator output.

Fix: update the wfextract directive to internal/domain/workflow and
regenerate all four extract surfaces via 'go generate ./...'. Re-apply
the documented hand-written nil-guards on interface wrappers (Yaegi
wrappers omit them natively), restating the IMPORTANT note in wfextract's
package doc (it now has interface wrappers for the first time). Delete
the stale hand-patched wfextract generated file in favour of the
correctly-named regeneration.

Regression coverage (the previously-missing types are now reachable from
interpreted code, proven by tests that fail against the old tables):
- wfextract/reachability_test.go: interprets a stage plugin referencing
  Trees/Reviewer/LoadPlaybookDirs and calls it; plus a Go-side nil-guard
  test on the three wrappers.
- pluginextract/reachability_test.go: interprets a daemon plugin building
  a Host from a Go-side Module and calls it end to end; plus nil-guard
  tests on the _Module/_Plugin wrappers.
- secretextract keeps its existing wrapper nil-guard tests.

Verified red-then-green: after a fresh 'go generate ./...' wipes the
guards, the guard tests fail with nil-pointer panics; after re-applying,
they pass. task check clean (113/113 UI, exit 0).
2026-09-03 03:30:23 +10:00
sam 8c19555b4a feat(playbooks): standalone lint tool for playbook binding dirs
Fourth slice of docs/prds/eda-playbook-engine.md (t2db.12): the design
doc's pre-merge collision layer, as a gopls-shaped standalone binary
(cmd/archie-playbooks) rather than a Taskfile wrapper -- one binary, one
schema source, mode-based entrypoints.

- cmd/archie-playbooks: subcommand dispatch table from day one (lint now;
  an lsp/serve mode later slots in without restructuring), flag parsing
  only per organisation.md's cmd/ rule.
- internal/app/archieplaybooks: the lint mode's composition -- collects
  directories, calls workflow.LoadPlaybookDirs (the daemon's own loader),
  reports findings. No second validation path: the linter agrees with
  runtime startup validation by construction.
- Reuses LoadKindWorkflowsYAML/LoadLabelWorkflowsYAML/LoadPlaybookDirs
  directly; collision/malformed/invalid bindings exit non-zero.
- Known limitation recorded: findings are file-granular, not
  line-granular (yaml.Unmarshal into map discards line numbers; a
  file:line diagnostic needs a yaml.Node upgrade, out of scope here).
- Taskfile lint:playbooks target; PRD records the shipped shape and the
  file-granular caveat.
2026-09-03 03:07:27 +10:00
sam cf6c5f42d5 feat(workflow): load kind/label bindings from multiple playbook directories
Third slice of docs/prds/eda-playbook-engine.md (t2db.11), completing open
question 2: instead of a single playbook dir, the config gains a LIST of
directories (playbook_dirs) of *.yaml/*.yml binding files, loaded at
startup as an additional input to workflow_routing_file and
workflow_labels_file, which remain supported unchanged.

- LoadPlaybookDirs(dirs []string) reads every yaml file across the
  configured directories and merges kind + arbitrary-label bindings.
- This is what finally exercises the design doc's drop-and-report
  collision rule: a single file cannot collide with itself, so the
  cross-source path (two files in one dir, two dirs, or a dir plus a
  single-file field) had never run in production. A binding key declared
  by more than one source is a reported load failure, not a silent
  pick-one by source order or version.
- Router-level collision handling: MergeKindWorkflows/MergeLabelWorkflows
  fail on a key bound by both the single-file fields and the playbook
  dirs, applied at boot.
- Config: PlaybookDirs []string, empty by default (no behavior change for
  existing configs). Repo-scoping (typing dirs to Config.Repos owner/name)
  is explicitly deferred to a separate later decision per ticket;
  org/tenant keying is out of scope.
- PRD: records the list shape per Sam's correction (single dir would
  contradict multiple independently-maintained playbook sources).
2026-09-03 02:57:38 +10:00
sam 17e626dfba feat(workflow): route arbitrary forge labels to workflows via playbook YAML
Second slice of docs/prds/eda-playbook-engine.md, complementing the
kind->workflow layer (WorkflowRoutingFile): the label vocabulary itself
becomes playbook data. A label outside the closed bug/feature/bootstrap
kind set (e.g. security) now dispatches to a registered workflow from an
optional WorkflowLabelsFile, instead of falling through to triage.

- LabelWorkflows + LoadLabelWorkflowsYAML/SetLabelWorkflows in
  internal/domain/workflow/routing.go, mirroring the kind layer's
  once-at-startup lifecycle. The closed Kind/NATS-subject set is
  untouched; the map only extends binding authority for labels the kind
  layer does not own.
- Collision rule per design doc: a label owned by the kind set, an empty
  label, an empty workflow name, or a duplicate binding is a reported
  load failure (dropped, logged, error returned to the caller) -- the
  schema defines what is accepted; nothing is silently arbitrated by
  schema/version.
- Route() precedence: explicit t.Workflow -> arbitrary-label binding ->
  kind binding -> triage -> implement -> default.
- Config: WorkflowLabelsFile (workflow_labels_file), wired at boot
  alongside WorkflowRoutingFile; documented in config.example.toml.
- PRD: records this slice's decisions (single-file shape, collision rule,
  precedence) so later slices stay anchored.
2026-09-03 02:44:41 +10:00
sam 5268634f27 feat(workflow): make kind-to-workflow routing YAML-overridable
First slice of docs/prds/eda-playbook-engine.md: which registered
workflow an intake Kind (bug/feature/bootstrap) prefers is now data,
loaded from an optional WorkflowRoutingFile, instead of a Go literal.
labelKinds and the NATS-subject-per-kind set are unchanged -- expanding
the label vocabulary itself is a separate, larger decision (subject
allocation) deferred per the design doc's open questions.

An unrecognised kind in the file is a reported load failure (schema
defines what's accepted, per the design doc's collision-handling rule),
not a silently-dropped binding. No file configured falls back to
today's hardcoded defaults unchanged.
2026-09-03 02:35:42 +10:00
sam ded0f0aef5 fix(store): satisfy makezero when building cipher payload
append(nonce, sealed...) appends to a slice initialized with a non-zero
length, which makezero flags. Build the nonce||ciphertext||tag payload with
an explicit buffer sized upfront instead.

Found by golangci-lint.
2026-09-03 00:01:11 +10:00
sam c13ec2f225 feat(webhook): harden captured payload redaction
RedactPayload only redacted values under a fixed set of sensitive key
names, so a sender that names a secret field something else leaked it into
captured_events.body. Harden the heuristic:

- broaden the key-name markers (compound names like access_token,
  client_secret, signing_key, x_api_key, webhook_secret, refresh_token)
  and keep gateway.sensitiveParameterKey identical to preserve the
  documented agreement between the two redaction paths.
- add value-shape detection: a compact JWT or a PEM private-key block is
  redacted under any key, catching the 'sender named the field something
  the heuristic misses' gap.
- keep it conservative so schema-by-example field mapping still works:
  only unambiguous shapes (eyJ-prefixed JWTs, PEM key blocks) are
  redacted by value, never ordinary content.

Best-effort per docs/prds/webhook-intake-security.md point 5.

Refs archie-core-t2db (captured_events.body redaction)
2026-09-02 23:59:43 +10:00
sam 7ee6ee5554 feat(store): encrypt playbook binding secrets at rest
binding.Binding.Secret was persisted as plaintext in the bindings.secret
column, so anyone with the SQLite file saw sender-side HMAC secrets. Seal
it with AES-256-GCM transparently inside the store: encrypt-on-write,
decrypt-on-read.

- cipher: AES-256-GCM, 12-byte random nonce, SHA-256-normalised 32-byte
  key resolved through the secret registry ([bindings].encryption_key).
- envelope 'arcie-binding:v1:<fingerprint>:<b64(nonce||ct||tag)>' so a
  future cipher/KDF change is a version bump and old rows stay readable.
- rotation via previous_encryption_keys; key identity is the material's
  SHA-256 fingerprint embedded in each row.
- legacy plaintext rows written before a key was configured stay readable;
  encryption is not retroactive.
- nil cipher (no key configured) preserves the pre-t2db.7 plaintext path.

Deferred: captured_events.body redaction upgrade (separate concern, see
docs/prds/binding-secret-encryption.md).

Refs archie-core-t2db.7
2026-09-02 23:54:06 +10:00
sam 96184a0e27 fix: resolve golangci-lint funlen, nilerr, and empty branch issues 2026-09-02 23:10:38 +10:00
sam 543c10ac9a feat(webui): playbook bindings -- matcher + mapping + workflow, HMAC-gated
Implements archie-core-t2db.4: the runtime-editable binding entity that
turns a captured webhook into an archie task, plus the per-source HMAC
verification at handleCapture that makes t2db.5's 'unauthenticated event
cannot trigger a binding' invariant structurally enforceable.

Captures, mappings and bindings are now separate concerns on the same
single *store.Store; the new BindingStore / BindingDispatcher /
BindingTaskCreator split keeps each interface under the interfacebloat
limit (8 methods).

State machine: draft -> pending_approval -> armed; any edit on an armed
binding drops it back to pending_approval, enforced in SQL via
status='pending_approval' on UPDATE so the rule cannot be bypassed by a
direct status field in the request body. Approve is the only sanctioned
transition into armed (WHERE status='pending_approval' guard).

Overlap: two bindings may not share a source in any status; enforced at
InsertBinding and UpdateBinding (transaction-local SELECT). Approve has
a separate TOCTOU guard so a race between Insert and Approve cannot
// bypass the check.

HMAC at handleCapture: reads X-Hub-Signature-256 (then X-Signature-256)
// fallback, verifies via webhookguard.VerifyHMAC against the binding's
// per-source secret, marks CapturedEvent.Authenticated=true on match.
// Captures with no armed binding, missing signature, or wrong signature
all persist as authenticated=false -- the dispatch loop's
binding.Matcher.Matches gate (requires authenticated) is what stops
them from spawning a task. Two armed bindings for one source returns
409 at capture time as a belt-and-braces TOCTOU guard.

Dispatch loop: added Daemon.dispatchBindings, called from both Cycle
and maintainAndDrain between poll and drain. Walks
captured_events where authenticated=1 AND source IN armed bindings
AND id NOT IN binding_dispatches, runs the binding's matcher, resolves
the mapping, enqueues a chat-style task with binding_id /
// binding_version provenance stamped in the same transaction as the
// row insert. The binding_dispatches ledger (composite PK on
(binding_id, capture_id)) is the at-most-once guarantee across cycles
// and daemon restarts.

Tasks schema migration: binding_id INTEGER NOT NULL DEFAULT 0 +
binding_version INTEGER NOT NULL DEFAULT 0 (taskSchemaVersion bumped
1 -> 2). Existing tasks keep 0 / 0.

Out of scope (documented follow-ups):
- secret encryption at rest (plaintext today)
- full dashboard HTML for the binding editor (CRUD API only this bead)
- per-binding rate limits (existing per-remote-address limit applies)
- binding.Owner / binding.Repo fields (dispatch falls back to the
  single configured repo; multi-repo deployments log and skip)
2026-09-02 23:09:25 +10:00
sam a4fbd8d6ef feat(webui): inline editing for per-repo allow_concurrent/max_retries/review_enabled
Adds PATCH /api/config/repos/{owner}/{name}, a dedicated seam distinct
from the generic dotted-key PATCH /api/config path: a repository is
one element of []config.Repo, not a flat key. Traced the mechanism
before adding any new plumbing -- overlay.Nest/ApplyOverlayValues
already handle "repos" as a single whole-slice key with no dot, so
the fix is a webui handler plus a bootstrap seam that reads the
daemon's live, untrimmed Repos (never webui.RepoView, which omits
Preflight/TestGlob), mutates the one matched field via the pure
applyRepoFieldUpdate helper, and republishes through the existing
UpdateConfig(ctx, {"repos": repos}) path unchanged.

config-row.js's row() gains an opts.save override so an edit can be
redirected to a different persistence call. The repositories table
renders an autosaving checkbox for the two booleans and a
save-on-blur/Enter input for max_retries, via api.configRepoUpdate.

Closes archie-core-b6ew.6, and with it archie-core-b6ew.
2026-09-02 22:35:03 +10:00
sam a444253209 test(webui): pin restart-required deliberateness and generic-render proof
TestConfigFieldDescriptorsRestartRequiredIsDeliberate pins the exact
RestartRequired value per field against the reload.go evidence
already documented in configFieldDescriptors' own comment, so a field
reclassified without re-checking reload.go fails loudly instead of
silently defaulting to "takes effect immediately."

ui/test/settings-schema.test.js renders settingsPage against a schema
fixture containing a field settings.js was never written to know
about, proving the renderer needs no field-specific branch to show it
correctly, and asserts by reading settings.js's own source that none
of the generically-rendered fields' keys appear literally in it --
the direct regression test for archie-core-b6ew.3.

The rest of this issue's scope (metadata completeness, structured/
locked non-editability, schema secret-leak guard) was already covered
incrementally while implementing archie-core-b6ew.1 through .4.

Closes archie-core-b6ew.5
2026-09-02 22:26:34 +10:00
sam 302caedec9 feat(webui): surface allow_concurrent/max_retries/review_enabled on the repositories card
RepoView.ReviewEnabled was previously absent from GET /api/config
entirely, not just unrendered; add it alongside AllowConcurrent and
MaxRetries. The repositories table gains Concurrent/Max retries/
Self-review columns as read-only pills -- this is the concrete
discoverability fix for "why isn't a second agent running": that
config was already reachable from the API, just never shown.

config.example.toml documents review_enabled, which had no example
entry before this.

Full inline editing of per-repo fields is out of scope here: the
dashboard's PATCH /api/config path has no array-index support for
updating one element of the Repos slice, which is separate,
real scope -- filed as archie-core-b6ew.6.

Closes archie-core-b6ew.4
2026-09-02 22:23:58 +10:00
sam 35eb4050de feat(webui): render scalar config sections generically from the schema
settings.js drops identityCard/budgetsCard/storageCard/webCard/
pullPolicyLabel in favour of one schemaCards() that walks cfg.schema
and builds a card per section, a row per non-structured field --
label, type, description, editability and enum options all come from
the backend descriptor instead of a hardcoded row(...) call per key.

config-row.js gains: enum fields edit as a <select> from opts.options;
opts.editable === false renders read-only distinct from a locked row
(no reason to show, just no edit affordance); opts.hint carries a
field's description as the label's tooltip, so "0 means unlimited."
travels with the field instead of being baked into a display-value
string per key.

ConfigSection also gains Description, filled in per section to match
each card's existing sub-text, since the card's own explanation is
schema metadata too.

Structured sections (repositories, models & providers) are untouched
-- still the existing bespoke renderers, reserved for archie-core-b6ew.4.

Closes archie-core-b6ew.3
2026-09-02 22:12:48 +10:00
sam e442eb5a72 feat(webui): attach the config field schema to GET /api/config
ConfigView.Schema carries the archie-core-b6ew.1 descriptor catalog
with each request's own values, locked reasons, and overridden
markers attached (buildConfigSchema), additive alongside the existing
flat fields so nothing that reads ConfigView today breaks.

Tests confirm schema values match their flat-field source of truth,
locked reasons propagate to the schema entry, structured fields stay
non-editable, and secrets never reach the schema block.

Closes archie-core-b6ew.2
2026-09-02 22:08:02 +10:00
sam 02c7829bee feat(webui): hand-author the configuration field descriptor catalog
Introduces ConfigField/ConfigSection and configFieldDescriptors(),
attaching key/label/type/editable/restart-required to every field
GET /api/config already exposes -- the data-independent half of the
schema contract from docs/prds/config-schema.md. RestartRequired is
sourced from reload.go's reloadableFields/reloadableSubFields
allowlist rather than guessed.

Not wired into the API response yet (archie-core-b6ew.2); this issue
is the catalog and its contract tests only.

Closes archie-core-b6ew.1
2026-09-02 22:05:46 +10:00
sam adf06e9efe fix(curator): stop curator actions from serializing a zero timestamp
session-memory's extracted actions never set Action.At, so the wire
carried time.Time's zero value ("0001-01-01") and the Curators
dashboard rendered it as "105694 wks ago". Stamp the action with the
curator's clock at the producer, and backfill any zero Action.At in
activityTracker.record from the pass's own timestamp so an engine
with no clock (skillcurator) can't hit the same bug.

Closes archie-core-z6wy
2026-09-02 21:52:28 +10:00
sam 7b1cef53eb fix(gateway): stop SOUL from forging the prompt invariant tags
Persona/SOUL text was interpolated raw at the top of archie.md.tpl,
so a hostile SOUL could inject its own <core_rules> or
<instruction_precedence> tags ahead of the real ones. Render it
through the existing xml-escaping helper into a labeled
trust="data" <soul> block, and tell instruction_precedence
explicitly that the block is reference data, not an authority.

Closes archie-core-1786637490097-12-bf415ffb
2026-09-02 21:48:57 +10:00
sam b147a288ae feat(workflow): dynamic triage instead of defaulting every task to implement
Route() had no content-aware fallback: any task with no explicit
workflow and no label match ran the full "implement" pipeline
regardless of content. Confirmed in production (task 6): a
chat-spawned task whose entire content was "this is just a test,
close it when you receive it" burned 669,421 tokens finding that out
the expensive way -- baseline, plan, and build all ran before the
plan stage's own conclusion (nothing needed to change) arrived too
late to save anything.

Adds a "triage" workflow: one cheap, read-only classification turn
(mirrors feasibility.go's decide-tool pattern) that either closes a
task needing no code change immediately, or sets Task.Workflow and
requeues under whichever workflow fits. Route()'s fallback now
prefers a registered "triage" over "implement", unchanged when
triage isn't registered. Labeled tasks are untouched -- a label is
already a free, reliable signal, so triage never fires there.

See docs/prds/dynamic-workflow-triage.md for the full design.
2026-09-02 21:40:06 +10:00
sam 9d36851b15 fix(workflow): net cache hits from token totals, trim gate-retry output
Confirmed in production (task 6, 2026-09-02): dashboard/PR summaries
showed raw prompt-token sums with no cache accounting -- 360,018 of a
reported 400,021 "prompt tokens" were cache hits billed at a steep
discount, not full price. Real fresh spend was ~40,003 tokens, roughly
10x less than the headline number implied. Nothing was double-counted;
the totals were internally consistent, just presented misleadingly.

Adds TaskContext.RunUsage (in-memory, presentation-only) and
formatTokenUsage, rendering "N tokens (F fresh + C cached)" in PR
bodies and the dashboard timeline instead of a raw sum, falling back
to the plain total when no breakdown is available.

Also trims StageBaselineGate's retry-mission prompt to failing-only
go test output via extractFailingGateOutput -- the full passing-
package listing was a direct, mechanical contributor to the growing
per-iteration prompt sizes on every gate-repair retry, not just a
display problem. Degrades to unchanged output for non-Go gates or
unrecognized shapes.
2026-09-02 21:28:49 +10:00
sam 52a460ebf9 fix(workflow): stop discarding baseline-fix commits as 'no changes'
Confirmed in production (carina, 2026-09-02, task 6): StageBaselineGate
spent 405,004 tokens genuinely fixing two broken pre-existing tests and
committed the fix locally, then the plan/build stages correctly found
nothing further to do for a content-free task. StageCommitPush decided
"no changes required" purely from the build stage's own empty diff,
with zero awareness that StageBaselineGate had already committed a
real, gate-verified change two stages earlier -- so it never pushed,
never opened a PR, and the whole worktree (including the real fix) was
discarded. 669,421 tokens burned to fix a real bug that then evaporated,
and it recurs: any task that hits the same red baseline pays the
identical tax again since the fix never lands.

Adds TaskContext.BaselineFixed, set when StageBaselineGate's own
CommitAll reports a real commit. The build stage's no-changes
determination now respects it, and StageCommitPush pushes whenever
either a fresh commit was made or a baseline fix already landed,
instead of erroring on an empty tree.
2026-09-02 21:25:54 +10:00
sam c7e3bf190e fix(channels): call MarkRunning once a channel is actually up
channels.Manager.MarkRunning was never called for telegram, email, or
webhook -- every start path only called MarkStarting then MarkFailed
on error, so the dashboard showed a working channel stuck on
"Starting" forever (confirmed live: Telegram polling and delivering
messages while /api/channels still reported "starting").

Adds gateway.Lifecycle{Starting, Running func()}, threaded through
Gateway.Start(ctx, router, lifecycle). Email and webhook report
Running once their listener actually binds -- webhook moves off
http.Server.ListenAndServe's internal bind to an explicit
net.ListenConfig.Listen so binding is observable before Serve.
Telegram reports Running after StartWebhook launches in webhook mode,
or after the first successful getUpdates response in long-poll mode,
observed via a pollReadinessClient http.Client wrapper since the
go-telegram/bot library keeps polling errors inside its own worker
goroutine. Restart re-reports Starting/Running each cycle.
2026-09-02 20:47:01 +10:00
sam cfc296f17f feat(image): define provider-neutral image capability contract and config
Adds internal/domain/image (Provider/Registry per the plugin engine
family shape, but no Lifecycle -- a provider is call-scoped, not
long-running) and config.ImageConfig with the same explicit
enabled=true cost gate MinimaxConfig already uses, since a hosted
provider spends real API credits and a local one can saturate the
host GPU. Design doc at docs/prds/image-capability-contract.md per
CLAUDE.md's rule for undesigned capabilities.

No provider implementation lands here -- that's the epic's next
child bead. Also re-links all 7 epic children that were unlinked
from archie-core-1786748942120-1-6636e629 (same orphaning pattern
found in the curator/memory epics earlier this session).
2026-09-02 20:20:54 +10:00
sam de3cd32c72 fix(logging): propagate bound slog attrs to wrapped handler in FeedHandler
FeedHandler.WithAttrs/WithGroup only tracked bound attrs for the live
feed copy; Handle forwarded the bare record to the wrapped handler, so
component (and any other With-bound attr) never reached stderr/file
JSON sinks or survived restart for historical log filtering. Now
WithAttrs/WithGroup also derive the wrapped handler via its own
With*, matching agentexec.SystemLogHandler's existing pattern.
2026-09-02 19:45:42 +10:00
samcharles93 7515ad82aa curator and other components aren't being logged or shown in the Web UI (archie) 2026-09-02 04:00:05 +00:00
samcharles93 70afbf7f95 fix: baseline gate repair (go test ./... -count=1) 2026-09-02 03:53:40 +00:00
sam c4871569ce feat(curator): session-memory curator reference implementation
Closes archie-core-1786637499114 (persistent session-memory curator),
the last open item in the curator engine epic.

docs/prds/session-memory-curator.md records the decisions this needed
before coding, most notably: Observation.Identity is the session ID,
not a user or cross-session identity. migration-decisions.md#5's
memory-scope mapping (global/agent-wide/user-wide/agent-user
relationship) is still explicitly open, and this issue has no business
deciding it unilaterally -- session-as-identity is the narrowest
unambiguous scope available today and is forward-compatible with a
later, properly-decided identity resolution.

curator gains ConversationSource (RecentSessions/Messages), gated in
Registry.validateDeclared/filter exactly like Skills/MemoryEngines, plus
Manifest.Conversations. internal/infrastructure/sessioncurator/adapter.go
adapts gateway.SessionStore's full CRUD/branch/search surface down to
those two read-only methods -- curator must not import gateway.Message,
so role derivation (message.From == botUser -> assistant) is copied from
gateway.compressTurnHistory's existing logic, not shared.

sessioncurator.go: Check is a cheap RecentSessions bool. Pass takes each
session's last 40 messages, asks the model for a JSON array of durable
facts (capped at 5/session), and writes each through
Registrar.MemoryEngines.Get(name) -- resolved by name at call time, the
same trust model every other curator capability already uses. A
response that fails to parse degrades to "nothing extracted" for that
session (not a Pass error); a real model-call failure propagates as one
(these were originally conflated into the same swallowed-error path --
caught by TestPassPropagatesALLMFailure, fixed by splitting the model
call from the parse step). PassInput.Since zero resolves to a 24h
lookback, not the beginning of time, so a cold-start first pass can't
read a session's entire history.

Manifest.OnInput = true wakes it via the existing WakeOnPrimaryInput
wiring (events.KindTurnCompleted), which already excludes every
curator-derived event kind from the forwarded set -- the trigger-
accounting invariant (archie-core-1786637489768) needed no new code to
hold here.

Wired into bootstrap.go:setupCurators via a new curatorLLMRunner adapter
(main.go) over the shared ai-sdk runtime, and sessioncurator.NewAdapter
over b.chatSessionStore/b.cfg.BotUser. Model resolves from
b.chatModels.ActiveModel() -- not b.defaultChatIdentity, which names a
task-routing identity, not a model reference (the same source
telegram_setup.go's sendChatTurn already uses for a real chat turn).

task check is green (full Go suite + dashboard tests).
2026-09-02 03:09:04 +10:00
sam c85bcff57d feat(curator): skill curator reference implementation
Part of archie-core-1786637499206 (skill curator reference
implementation), under the curator engine epic.

docs/prds/skill-curator.md records the design decision this needed
before coding (no settled design existed for what "review, prune, and
improve skill definitions" means as concrete Pass() behavior): v1 is
mechanical validation and safe whitespace normalization only. No
LLM-authored rewrites of skill content, no automatic deletion -- an
autonomous process editing or deleting a user's skill has no undo path
today, and the epic's actual purpose for this issue is proving the
curator contract holds for a second, structurally different consumer
(files-as-skills vs. observations-as-memory), not solving "what makes a
skill good."

internal/infrastructure/skillcurator/store.go: a curator.SkillStore
implementation over one root's .agents/skills/*/SKILL.md files. Not
internal/skill.Discover directly -- Discover parses every skill up front
and fails the WHOLE call if any one of them doesn't parse (correct for
"load the catalog to boot the daemon", wrong for "review each skill and
report what's wrong with it individually"). This store lists/reads raw
bytes per skill; the curator itself parses each one so one bad skill's
parse failure becomes one Action, not an aborted pass.

internal/infrastructure/skillcurator/skillcurator.go: the CuratorEngine.
Check is a cheap directory-listing bool. Pass classifies each skill into
exactly one of: skill.invalid (frontmatter parse failure, left
untouched), skill.incomplete (missing required name/description,
untouched), skill.normalized (trailing whitespace / missing final
newline, written back), or nothing (already clean). Manifest declares
only Skills -- no Tools, no MemoryEngine -- since nothing here needs
either.

Wired into internal/app/archied/bootstrap.go:setupCurators, rooted the
same way loadWorkflows already resolves skillsBase (cfg.SkillsDir falling
back to cfg.WorkDir) -- one local root, not the full multi-root catalog a
chat turn reads. Skills is now a shared registrar-level service
(alongside Events/MemoryEngines), narrowed out of a curator's view by
registry.filter when undeclared, same as every other capability.

task check is green (full Go suite + dashboard tests); go test ./...
is green.
2026-09-02 02:58:36 +10:00
sam 704a5ee642 fix(config): reject memory.provider instead of silently ignoring it
Closes archie-core-1786637499161-356-e424e40d.1.

setupMemory's own doc comment claimed external providers are added via
RegisterExternal when cfg.Memory.Provider is set -- no such call
existed. Investigated wiring it for real: there is no external
MemoryProvider implementation anywhere in the repo to construct and
register (only internal/memory/builtin.Provider exists), so there is
nothing to route memory.provider's value to yet. Wiring the call would
be dead plumbing to nothing, not a fix.

The actual defect is that setting memory.provider has always silently
done nothing -- an operator who set it reasonably believed they
configured something. validateMemory now rejects it outright with a
clear reason, turning a silent no-op into a loud one at config load,
same failure-loudly posture as every other unimplemented-value check in
this file. Drop the rejection once a real external provider (or the new
domain/memory engine family's Honcho proof, archie-core-1786637492031)
gives the field an effect again.

Updated setupMemory's doc comment and config.example.toml to match --
the provider_config example is removed too, since it has nothing left
to configure.
2026-09-02 02:50:35 +10:00
sam 388b093b78 feat(curator): bind curators to memory engines
Closes archie-core-1786637499636 (bind curators to memory engines).

curator.Registrar gets MemoryEngines, a MemoryEngineSource
(Get(name) (domainmemory.MemoryEngine, bool)) mirroring
domain/memory.Registry.Get's exact signature so the real registry
satisfies it with no adapter. Registration enforces presence the same
way Tools/Skills already do: a curator declaring Manifest.MemoryEngine
with no MemoryEngineSource on the host is rejected at Register, and
filter() clears MemoryEngines from a curator's view when it declared no
engine. A curator with a declared name that turns out not to exist in
the registry is not this layer's problem -- Get(name) at call time
returns false, same as ToolBuilder.Build with an undeclared tool.

Resolution is by name at call time, not a value pre-fetched at Bind: a
curator calls MemoryEngines.Get(its own Manifest().MemoryEngine) itself,
the identical trust model ToolBuilder.Build(ctx, declared) already
uses -- the curator is trusted to pass its own declared identifier, not
handed something pre-scoped to it. Swapping the configured engine (a
different name resolving through the same registry) changes where a
curator's memories land with no change to the curator itself, since
nothing about the curator's own code names a concrete engine.

setupCurators wires b.memEngines (*domainmemory.Registry, set by
setupMemoryAll which Run() calls first) straight in -- it already
satisfies MemoryEngineSource.

A curator declaring no memory engine remains valid (existing behavior,
now also covered by the "nothing declared" and "no source" table
cases) and is only ever affected by a memory operation it never has the
means to attempt, since MemoryEngines is nil in its view.
2026-09-02 02:47:48 +10:00
sam 538b29aa0f feat(daemon): wire the domain/memory engine registry at boot
Completes archie-core-1786637499480 (ship the file-writing store as
the default memory engine): cfg.Memory.Engine now actually selects and
starts an engine, closing the gap the previous commit's message called
out as remaining.

setupMemoryEngine builds a domain/memory.Registry, registers
infraMemory.BuiltinEngine for "builtin" (and the empty/unconfigured
case, matching applyMemoryDefaults' resolution) rooted at
workDir/memory-engine -- separate from the legacy provider's
workDir/memory, so the two can never collide on the same files while
both exist -- starts the registry, and stops it on daemon shutdown.
cfg.Memory.Engine is already validated against the same set at config
load (configuration.validateMemory), so the switch's default case only
guards the two lists drifting apart, not live operator input.

setupMemory and setupMemoryEngine are both called through a new
setupMemoryAll from Run(), which was already at its cyclomatic
complexity budget -- a genuine grouping (both are memory-subsystem
startup), not two unrelated steps merged just to hit a number.
setupMemoryEngine deliberately does not thread Run()'s ctx through to
registry.Start, matching setupMemory's own existing choice for the same
reason (Start is a synchronous startup step, not a subscription that
needs the daemon's lifetime the way setupCurators' WakeOnPrimaryInput
does) -- also what avoids a contextcheck trip that threading ctx through
setupMemoryAll produced during development.

Not yet done: internal/memory.Manager (the live system-prompt/tool-call
path) does not read through this registry -- the two memory systems run
in parallel, per archie-core-1786637490069-11-e245c170's explicit "do
not extend the legacy system in place" scoping. That convergence is
future work, not part of this issue.
2026-09-02 02:44:12 +10:00
sam f3abcb859e feat(config): config surface for memory engine selection
Part of archie-core-1786637490043 (config surface for memory engine
selection and settings).

config.MemoryConfig gets an Engine field selecting among the
domain/memory engine family, plus toml tags on every field -- Provider,
ProviderConfig, and SessionTTL had none before, yaml/json only, which
didn't match repo convention.

Engine and Provider are kept as two separate fields rather than one:
they select between two different mechanisms (the new engine family vs.
the legacy external-provider path, archie-core-1786637499161-356-e424e40d.1),
and conflating them would mean a value meant for one silently getting
read by the other once both are wired.

applyMemoryDefaults resolves an absent engine to "builtin" so
unconfigured and explicitly-builtin are the same state everywhere after
defaulting. validateMemory rejects an unrecognised engine name in the
same style as forge.type's existing validation -- named constant list,
"%w: field %q (want %s)" shape, wired into the same validate() chain.
config.example.toml documents the new field.

Not yet wired to anything: nothing reads cfg.Memory.Engine at daemon
boot yet to actually select and start an engine through
domain/memory.Registry. That's bootstrap wiring, tracked as the
remaining half of archie-core-1786637499480 (ship the builtin store as
the default engine).
2026-09-02 02:40:10 +10:00
sam a5d3472b4b feat(memory): builtin engine implementing the domain/memory contract
Part of archie-core-1786637499480 (ship the file-writing store as the
default memory engine).

Records the missing design decision first (migration-decisions.md#5,
"authoritative memory record and revision model" was explicitly open):
internal/memory/builtin.Store addresses blocks by content substring with
no stable per-block ID, which the new Record{ID}/Forget(id) contract
needs. Decided and implemented: an ID assigned once at Write via a
leading `<!--mem:<uuid>-->` marker line, stripped back out of
Record.Content on read, located by that marker (not content) for
Forget -- so an unrelated edit can never orphan a Forget. Identity maps
to a directory named hex(sha256(identity)), not the identity string, so
an arbitrary identity can never traverse outside the configured root.
Kind maps to a builtin section name via the store's own
ValidateSectionName, so an invalid Kind fails the same way an invalid
section already does.

internal/infrastructure/memory.BuiltinEngine wraps one builtin.Store per
identity and implements domain/memory.MemoryEngine end to end
(Write/Query/List/Forget, Lifecycle, Manifest). Lives under
internal/infrastructure per organisation.md ("infrastructure implements
domain contracts"), not internal/domain/memory, which stays
contract-only.

internal/memory/builtin is untouched -- confirmed via git status and its
own test suite passing unmodified, per this issue's acceptance
criterion. This engine is not yet wired into daemon bootstrap or the
config surface (archie-core-1786637490043); it proves the contract
holds for the builtin store the same way curator's reference
implementations prove the curator contract, and registers/runs cleanly
through domain/memory.Registry (TestBuiltinEngineRegistersAndRuns
ThroughTheRegistry).
2026-09-02 02:36:10 +10:00
sam a500603e27 feat(memory): typed memory engine contract, registry, and registrar
Part of archie-core-1786637490069-11-e245c170 (memory engine contract
and owning registry), under epic archie-core-1786637499161 (pluggable
memory engine).

Stands up internal/domain/memory as a conforming engine family per the
strict plugin engine rule, mirroring internal/domain/curator's split
exactly as that issue's notes specify:

- contract.go: MemoryEngine (Lifecycle + Store + identity + Bind), with
  Store carrying the real domain operations the epic requires --
  Write/Query/List/Forget -- not just Name/Version. Store is split out
  from MemoryEngine so the interface stays under golangci-lint's
  interfacebloat threshold the same way curator's embedded Lifecycle
  does: an embedded interface counts as one slot, not its expanded
  method count.
- registrar.go: narrow typed host access (EventSink, Clock). No
  capability-gated fields yet -- unlike curator's Tools/Skills, no
  memory-specific host service exists to gate on today, so Registry.Bind
  passes the host straight through rather than filtering.
- registry.go: Register/Get/Names/Start/Health/Stop with the same
  duplicate/nil/panic/failure-isolation guarantees as curator's registry,
  ctx-bounded Stop, reverse-order shutdown.

This is the contract only -- no engine implements it yet, and
internal/memory's existing ad-hoc MemoryProvider/Manager (pre-migration)
is untouched, per the issue's explicit "do not extend in place" note.
2026-09-02 01:55:11 +10:00
sam b966f6bd03 fix(telegram): send undelivered-media fallback after finalize as its own message
Closes archie-core-atv2.

Media delivery is fire-and-forget with a 30s budget, far longer than
the turn's text usually finishes -- so a large upload failing (or
timing out) after doFinalize has already retired the renderer goroutine
is the normal case, not an edge one. appendFallbackLine was still
writing into toolLines and calling requestRender, a non-blocking send
into a channel nothing reads any more once stopRendering has returned:
the line was silently dropped and the finalized message stood
unchanged, with nothing telling the user or the agent the file never
arrived.

stopRendering is the single point every finalize/abandon path funnels
through to retire the renderer, so it's the right place to mark the
reply finalized. appendFallbackLine now checks that flag and, once
set, sends the fallback as its own follow-up message instead of
appending to a buffer nothing will render again.
2026-09-02 01:18:54 +10:00
sam f603305a6a fix(telegram): stop adjacent rich blocks from rendering glued together
Two bugs behind the "System statusEverything is operating normally."
render:

1. Consecutive plain lines with no blank line between them were joined
   into one paragraph's text with an embedded "\n". Telegram's
   rich-block paragraph does not render an interior "\n" as a break --
   it renders nothing, so the words ran together with zero separator.
   Join with a space instead: the correct rendering of a CommonMark
   soft line break anyway, and it survives regardless of how the
   client treats a literal "\n".

2. Telegram's renderer adds no vertical gap between adjacent blocks on
   its own -- only a heading carries its own margin -- so a label line
   immediately followed by a list (or any two non-heading blocks back
   to back) rendered touching, with the list's first item glued to the
   preceding text and no visible line break at all. appendBlock now
   inserts an empty spacer paragraph before any block that doesn't
   follow a heading, making the gap explicit instead of depending on
   client-side spacing that isn't there.

Reproduced directly against markdownToBlocks with the reported input
and confirmed the fix before writing this up.
2026-09-02 01:12:53 +10:00
sam 19a16e8c63 fix(update): drop dev-facing wording from the unconfirmed-update message
"Treat this as UNCONFIRMED, not complete" is developer-speak in a
message an operator with no codebase context has to act on. Reword to
plain instruction: say what's known, then what to do about it.
2026-09-02 00:35:13 +10:00
sam 01a1d1ba21 fix(logging): make Tail/Page allocation bound explicit for CodeQL
CodeQL's Go analyzer (go/uncontrolled-allocation-size, alert #6) can't
trace min(limit, MaxTailLines) as a bound on the request-derived limit
before make([]Entry, 0, limit). The clamp was already correct and has
been in place since 48133e6 -- confirmed by rescanning at current HEAD,
the alert still fires -- so this is an analyzer gap, not a live bug.
Same clamp, written as an explicit if so static analysis (and readers)
can verify the bound directly.
2026-09-02 00:04:06 +10:00
sam 1425050506 fix(build): stamp gatewayVersion/runtimeVersion in the right package
-X main.gatewayVersion/main.runtimeVersion targeted package main, but
those vars live in internal/app/archied (cmd/archied is a thin
wrapper). An -X target that doesn't exist is silently ignored by the
linker, so every archied binary this shipped -- task build, install.sh,
Dockerfile.archied, and the carina self-update script -- reported
"dev" instead of its real version. Point every one of those ldflags at
the actual package.

Also, while chasing the carina incident this surfaced:

- restart.go: /restart didn't check for an in-flight /update install.
  The install runs as archied's own child process, so a restart killed
  it mid-build (observed: "signal: terminated") instead of letting it
  finish or fail cleanly. Refuse the restart while updateInProgress.
- update.go / update_report.go: the "no version could be confirmed
  running" branch still opened with "Update complete", contradicting
  its own next clause. Reworded so an unverified outcome isn't
  reported as completion.
2026-09-02 00:00:32 +10:00
sam 95342037b0 test: replace brittle assertions with behavior checks
- mcp client_test: assert both initialize handshake calls happened and
  notifications/initialized has an empty ID, not their exact order
- telegram model_test: assert the resolved model has an openrouter/
  prefix instead of scanning button text for the string; replace the
  single-hardcoded-count pagination test with cases covering 0/1/8/9/100
  models so pageBounds and nav-row logic are actually exercised
- webui api_chat_test: assert the actionable file path is present in the
  streamed error instead of matching generic wording
2026-09-01 23:39:23 +10:00
sam 86ccff8787 chore: nolint for containedctx 2026-08-31 23:02:14 +10:00
sam ae33e366b2 chore: errorlint cleanup 2026-08-31 22:58:37 +10:00
sam 1adb018d3f chore: noctx cleanup 2026-08-31 22:56:38 +10:00
sam bd0867bd05 chore: remove unusued list 2026-08-31 22:54:11 +10:00
sam 7ff008e89d test(telegram): tidy install-context detachment regression test 2026-08-31 22:53:53 +10:00
sam 344c2d3491 test(webui): replace brittle origin response assertions 2026-08-31 22:53:16 +10:00
sam 14d3a9461f feat(workflow): adversarial self-review stage before PR open
Implements archie-core-h019.3 (reviewer isolation) and h019.4 (wiring),
per docs/prds/adversarial-self-review.md.

- internal/domain/workflow: new Reviewer interface and StageReview,
  inserted between StageDiffCap and StageOpenPR in the implement
  workflow. Fail-closed: a review that didn't run, or a confirmed
  error-level finding, parks the task (store.StatusParked) with
  findings in Detail; warn/plausible findings never block. Inert
  unless Repo.ReviewEnabled.
- internal/worktree: Manager.Snapshot exports HEAD's tracked files into
  a fresh directory with no .git, for the reviewer's isolated
  workspace. Trees interface gains Snapshot; hybridTrees delegates to
  it.
- internal/app/agentworker: subagentReviewer implements Reviewer using
  ai-sdk's agent.Subagent (nested, in-process, non-streaming
  generation) -- conversation isolation is structural (Subagent.Run
  never carries message history), workspace isolation is the snapshot
  above. The reviewer's toolset is read-only (read/grep/find) rooted
  at the snapshot, never the worker's own toolset, so it cannot reach
  the worktreerpc publication grant. Findings are captured via a typed
  record_finding tool appending to a slice the stage owns, read even
  when Subagent.Run errors (a truncated run's captured findings
  survive).
- internal/config: Repo.ReviewEnabled (default false) and
  Models["reviewer"] (falls back to Models["builder"], no schema
  change).

Judgment calls: Reviewer.Review never returns a Go error -- every
failure folds into ReviewReport per the PRD's fail-closed contract,
which the stage can act on uniformly. A Subagent.Run error with zero
captured findings resolves to ReviewStatusNotRun (park); a
Subagent.Run error with >=1 captured findings resolves to
ReviewStatusCompleted with those findings, since the PRD's own example
("three defects survive a step-limit truncation") requires them to
still be evaluated for blocking rather than discarded.
2026-08-31 21:51:33 +10:00