Backlog hygiene per the file's own maintenance rule — resolved items get
moved out, not left ticked-in-place.
- Cut resolved body bullets from BACKLOG §A/B/C/D/E/G/I (health probes,
static-first infra, CritiqueFinding, architect contradiction-check, A1/A2
gates, batch fetch-approval, source/egress events, render matrix, markdown,
session list, idea promotion, turnId/probe/narration, .cs symbols, scene
validator). All already archived in RETRO's burndown table.
- Narrow partially-done items to their live follow-up rather than delete:
B5 -> static_check stage emitter only; B6/C-A3 -> CritiqueOutcomeCorrelated
producer only; C-A1 -> planner-prompt activation only.
- Drop the stale "§E UNBLOCKED" index line (plan-comparison is BLOCKED, no
PlanCandidate event; idea-promotion shipped f107ff5).
- Archive the operator-requested TUI wave (palette, @ picker, CLAUDE.md L0
injection, inline action rows, free-scroll/help/filter) in RETRO 2026-06-22
— never BACKLOG items, would otherwise be orphaned in neither file.
- gitignore apps/tui-go/preview (cmd/preview build artifact).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Three TUI-polish items.
- OUTPUT free-scroll: PgUp/PgDn (and ^u/^d half-page) scroll the transcript
back through history; esc snaps to tail-follow. outputScroll is clamped
against outputViewport(), which mirrors View/renderMain geometry, so the
edges land exactly (no dead presses unwinding an overshoot). routerRows
split into buildTranscriptRows (full render) + windowing so the handler
measures the same content. Tests cover the clamp + the fits-in-panel no-op.
- Help overlay: `?` (or the "help" palette command) opens a keybinding
cheat-sheet grouped by context (session / overlays / approval band) — the
non-palette keys especially. Any key dismisses it.
- Event-inspector filter: `/` filters the event list by a case-insensitive
substring of type/detail (currentEvents applies it); `c` clears, esc stops
editing then closes. Fresh per open. EVENTS side panel stays unfiltered.
go build / vet / test green; rendered help + filtered inspector via
cmd/preview (help, events-filter kinds).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The ⏵ tool-start row doubled up with the ✓/✎ result row in OUTPUT. Drop it
inline — tool starts still show in the tool list + EVENTS panel. The inline
transcript now carries only outcomes: ✎ writes, ✓/✗ tool results, ⌘/✕
approvals, ⊞/⊟ grants.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Surface side-effecting ("external feedback") events in the conversation
flow, opencode-style, instead of only in the EVENTS side log.
- New RouterEntry role "action" (glyph + text) appended at arrival so it
interleaves with chat turns by order. Rendered dim with an accent gutter
glyph; the EVENTS panel still carries the full stream.
- Surfaced: tool start (⏵), tool done (✓ / ✎ wrote <path> (+a −b) for a
diff), tool failed (✗), rejected (✕ blocked), approval resolved
(⌘ approved / ✕ rejected, noting "· via grant" on a grant auto-approve),
and grant create/revoke (⊞ / ⊟ · scope). ApprovalResolved names no tool
on the wire, so the tool is recovered from the pending queue before the
gate is dropped.
- Toggle: "inline actions" palette command flips actionsHidden (rows are
always recorded, gated at render, so toggling reveals history); persisted
in tui-prefs.json. prefs.go refactored to load/mutate/save so the new
field and statusbarHidden no longer clobber each other.
- demo.go: "actions" preview kind.
go build / vet / test green; rendered the flow via cmd/preview.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Surface the new cross-session grant scopes in the TUI.
- Approve-always (A) now opens a scope picker instead of immediately
creating a SESSION grant: choose this session / this project / everywhere.
SESSION stays the default (A then enter/s = old behaviour); p and g
create PROJECT / GLOBAL grants. The grant + approval are sent on confirm
(esc cancels, leaving the call pending).
- New "grants" palette command + G shortcut open a standing-grants viewer
(OverlayGrants) listing the active PROJECT/GLOBAL grants — scope, tool,
permitted tiers, project path — with x/enter to revoke. The server's
RevokeGrant reply (a fresh grant.list) refreshes it in place.
- protocol.go: TypeGrantList + GrantDto; RevokeGrant/ListGrants encoders;
CreateGrant now documents the wider scopes. server.go decodes grant.list
into m.grants and treats it as a non-session global reply.
- render-matrix coverage: grant.list classified as a non-rendering query
reply (populates the overlay, not the transcript).
- demo.go: grants + grant-scope preview kinds.
go build / vet / test green; rendered both overlays via cmd/preview.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Make the grant system's wider scopes real and add a revoke producer.
Before: grants were loaded per-session (the gate folds only the running
session's own stream), so GrantScope.PROJECT — though declared and matched
in the engine — was dead in practice, and there was no GLOBAL scope or any
way to revoke a grant (ApprovalGrantExpiredEvent had no producer).
- GrantScope: add GLOBAL(toolName); make PROJECT tool-bound. Every
operator-creatable scope is now tool-bound so a grant can never be a
blanket "approve everything" (that's YOLO mode).
- DefaultApprovalEngine.scopeMatches: GLOBAL matches the bound tool in any
context; PROJECT matches when the session's projectId AND tool match.
- Cross-session ledger: PROJECT/GLOBAL grants are appended to a reserved
GRANT_LEDGER_SESSION_ID stream instead of a session's. The approval gate
now unions the ledger's grants with the session's, and derives projectId
from the bound workspace root (ProjectIdentity.of) so PROJECT grants match
later sessions on the same repo. SESSION/STAGE grants still live in (and
die with) their session.
- Revoke: ClientMessage.RevokeGrant appends ApprovalGrantExpiredEvent to the
ledger (reducer already drops it); ListGrants/GrantList expose the active
standing grants for the TUI viewer.
- Drop the server-side T2 grant ceiling per operator request: a grant may
now authorize any tier (incl. destructive T3/T4). Tool-binding is retained
as the remaining guard.
Backend only; the TUI scope picker + grants viewer follow. core:events,
core:approvals, core:kernel, apps:server compile; approvals/events/kernel/
server suites green; new GrantScopeMatchingTest (5) covers GLOBAL/PROJECT
match, project isolation, and the no-cap T4 path.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Palette: each command now shows its bare-key shortcut in a key column (and the filter
matches on it), so the palette teaches the shortcuts instead of hiding them.
Status bar: a new "status bar" palette command opens a toggle overlay to show/hide the
non-essential segments (current stage, session status, workspace path, model, last-event
clock, resource gauge); correx/connection/session-name/spinner stay. Choices persist
TUI-local in tui-prefs.json (JSON in the config dir — display state, kept out of the
shared/replayed server config) and reload on launch.
Also trims the in-session footer (drops stats/model — both reachable via `p cmds`) so it
no longer overflows + truncates `q quit` at ~100 cols after the transcript-nav additions.
Tests cover toggle→render gating and prefs persistence; verified via cmd/preview renders.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
On session start, discover CLAUDE.md and AGENTS.md at the bound workspace root and inject
their (concatenated, header-labeled) contents as an L0 system context entry for every stage —
mirroring the .correx/project.toml ProjectProfile path end to end. Recorded as
AgentInstructionsBoundEvent (invariant #9) so replay reads the recorded fact, not the live
file; folded into SessionState.boundAgentInstructions and rendered by buildAgentInstructionsEntry
right after the project profile. AgentInstructionsLoader reads root-only, both files if present.
Bound via ServerModule.bindAgentInstructions alongside bindProjectProfile.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Typing `@` at a word boundary in the chat input (mid-word `@` stays literal, e.g. emails)
opens a workspace file picker: it requests file.list for the session (cached per session),
filters as you type, and enter inserts `@path ` at the cursor — back to typing. Renders as
a transparent overlay, windowed around the selection. file.list is classified non-rendering
in the render-matrix guard. Tests cover token-boundary detection, ref insertion, and filtering.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
New WS query: ClientMessage.ListFiles{sessionId} -> ServerMessage.FileList{paths}
(@SerialName "file.list"). StreamQueries.listFiles resolves the session's bound
workspace root (latest SessionWorkspaceBoundEvent) and lists it via WorkspaceFiles —
a path-only walk that reuses RepoMapIndexer's directory-ignore convention but, unlike
the indexer, never reads file contents and includes every file type (configs/assets,
not just source+markdown), sorted and capped. Tests: WorkspaceFiles (types/ignore/cap/
missing) + file.list wire round-trip. Go @ autocomplete consumes this next.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
ctrl+↑/↓ jumps between your sent (user) messages in the transcript: ctrl+↑ from the
bottom selects the most recent, steps to older ones (clamping); ctrl+↓ steps back and
drops to tail-follow past the last. The selected message is highlighted and scrolled
into view (routerRows tracks per-message row offsets and windows to the selection).
`y` yanks the selected message — or the latest turn when none is selected — to the
system clipboard over OSC52 (go-osc52, tmux-wrapped under $TMUX), which is SSH/Termius-
safe unlike a mouse drag. A brief "✓ copied" footer note animates out (and self-stops
the tick). esc drops the selection; switching sessions clears it. Tests cover the
user-message jump/clamp/tail and the copy-text selection.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Modals now composite over a dimmed copy of the screen behind them instead of an opaque
scrim, so the transcript stays faintly visible around a modal (opencode-style). center()
stashes the frame's base (View sets m.lastBase), strips+dims it to a faint scrim, and
splices the modal box over it (ANSI-aware via ansi.Truncate/TruncateLeft so the side
margins keep the dimmed content). An idempotence guard passes an already-full-screen
modal through unchanged, so the existing double-center builders don't get re-dimmed.
Geometry tests assert the composite still fills the screen exactly.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Typing `/` as the first char of a chat line opens the command palette inline
(opencode-style), instead of typing a literal slash; mid-line `/` stays literal.
Factors openPalette() (shared with the `p` key) and advertises `/ cmds` in the
insert-mode footer when the line is empty.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Input history is now a single global ring (recordInput) shared by every input surface —
free-chat, the workflow-intent line, and in-session chat — so ↑/↓ recalls anything you've
sent, including outside a session (previously per-session, so idle/intent had none).
Copy fix: the 120ms animation tick now only runs while something actually animates
(animating(): active session spinner, blinking caret, reconnect). Init no longer starts it
unconditionally; tickKick (re)starts it on demand and tickMsg stops it when idle. An idle
screen no longer auto-redraws, so a native terminal text selection survives instead of being
wiped every frame. Tests cover the ring (dedup/cap/cross-context walk) and the idle gate.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
64a90e7 gated the probe on a static [[providers]] llamacpp url, missing the primary
managed path (sample-config.toml's [[models]], where correx spawns llama-server at
[models].host:port and registers no provider entry). Fall back to that host:port when
[[models]] is configured so the LLAMA_SERVER subject is monitored on the common setup.
Surfaced while drafting the A§4 live-QA plan. Compile green.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Wire the built-but-unregistered LlamaServerHealthProbe (266bbf0) into the health monitor:
when a `llamacpp` provider URL is configured, add an HttpLlamaLivenessClient-backed probe
(dedicated CIO client, 3s per-ping timeout, shutdown-hook close) to the probe list. Gated
on a configured provider URL so a model-less box never reports a spurious LLAMA_SERVER
DEGRADED. Replaces the Main.kt TODO(wiring). main() is not unit-tested; the probe + client
are already covered by 266bbf0 — compile + :apps:server:test + detekt green.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Wire the display-only ArchitectContradictionChecker (eae0a0c) live: ServerModule.start()
subscribes to the architect stage's `design` ArtifactCreatedEvent, resolves the decision
text (prefers the design schema's `approach` field; falls back to the capped artifact
text), runs checker.check(...), and appends the non-null PossibleContradictionFlaggedEvent.
Live-only and never halts a stage (runCatching). Checker built in Main.kt gated on
project.enabled (same L3 "project:" namespace ProjectMemoryService.persist populates) and
threaded into ServerModule as a nullable param. Replaces the standing TODO(wiring) in Main.kt.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Promote an auto-captured idea from the cross-session board into the curated per-repo
profile at <workspaceRoot>/.correx/project.toml as a `conventions` entry. New
IdeaPromotedEvent tombstones the idea off the board (IdeaReader treats it like a
discard) and records the promotion as a fact; the server PromoteIdea handler loads
the profile, appends the idea text (deduped via ProjectProfile.withConvention), and
writes it back. Adds ProjectProfileWriter (escape-aware TOML serialize/write) and
makes SimpleToml's reader unescape \\ and \" symmetrically with the writers
(fixes a pre-existing read/write asymmetry surfaced by the round-trip test).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
POST /sessions/{id}/approve-sources extracts distinct hosts from a source list
(SourceListHosts.extract) and emits one EgressHostsGrantedEvent for the session, so
the live per-session egress allowlist auto-clears subsequent web_fetches to those
hosts — approve the list once instead of per-URL.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- pending approvals are a navigable queue (PendingQueue/PendingIdx; ↑↓/jk, count
shown), no modal stacking; resolve removes the specific gate by request id
- y/n/e keys (approve/reject/steer); T0–T2 act on one key, T3+ requires arm+confirm
("press y again"), any other key disarms; reject/steer/auto never gated
- snapshot restore now rehydrates the full pending list, not just the first
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Table-driven render assertions (37 cases over 42 protocol Type* constants), driven
through applyServer and asserted ANSI-stripped on stable defining text. Coverage
guard parses every Type* from protocol.go and fails if a kind is neither covered
nor explicitly classified non-rendering. Pins width/theme/timestamp; no behavior change.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
R opens a navigable overlay of recent sessions (GET /sessions: short id, status,
workflow/stage, relative age); enter resumes via POST /sessions/{id}/resume, which
replays onto the existing global /stream (no WS re-dial). Fetch/resume errors surface
in the overlay, never panic. stdlib-only (ws.Client.HTTPBase derives the base URL).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
renderMarkdown wraps charmbracelet/glamour (dark style, per-width memoized) and is
hooked into the router chat-turn render path; falls back to plain content on any
glamour error so the TUI never crashes or loses text. Other roles/UI untouched.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
`go 1.26` is not a fetchable/released toolchain, so released Go cannot build the
module ("toolchain not available"). 1.24.2 is the minimum the deps require
(charmbracelet/x/ansi needs >= 1.24.2) and is a real toolchain; the directive is a
floor, so newer Go still works. Bump if a specific newer release is intended.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Pause narrations are tagged with a pauseKey; a per-session pending-pause set gates
the lane worker so a resolved/resumed pause is skipped instead of blending with the
current status. ApprovalDecisionResolved drops the specific request's pause;
OrchestrationResumed clears all session pauses. Replay-safe (no recorded events change).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- PossibleContradictionFlaggedEvent + RelatedDecision (registered + serialization test)
- ArchitectContradictionChecker: L3 similarity retrieval over prior decisions,
threshold-gated, fake-testable; non-blocking (surfaces candidates only)
- live wiring left as documented TODO (needs an architect-decision emit hook +
in-session decision embedding into L3)
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Conservative type-declaration pattern (class/interface/struct/enum/record);
no method capture to avoid false positives. (BACKLOG I)
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- EventStoreLatencyProbe: times a side-effect-free lastGlobalSequence read,
wired into HealthMonitor; HealthConfig.eventStoreWarnLatencyMs (BACKLOG A §4)
- LlamaServerHealthProbe: liveness + tokens/sec degradation trend, injectable
client (HttpLlamaLivenessClient over Ktor); deterministic unit tests
- llama probe left unregistered (TODO) — HttpClient not in monitor scope yet
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- repomap: tag prefix match requires trailing ':' so /repo can't match /repo2
(L3RepoKnowledgeRetriever filter + ProjectMemoryService write; the existing
existsByTurnIdPrefix check already included the delimiter)
- WorkspaceStateProbe fingerprint joins entries with newline to match docstring
- regression test for /repo vs /repo2 non-collision
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The repo-map turnId tag and the L3 retriever filter used 'repomap:$repoRoot' with no
delimiter, so a retriever for /repo also matched /repo2 entries (startsWith). Add a trailing
':' to the tag (no-stateKey branch) and the retriever filter so the repoRoot boundary is
explicit. Test: /repo retrieval no longer leaks /repo2 entries, still matches its own
with/without a stateKey suffix.
First slice of the plan-generation pipeline epic. A pure-Kotlin PlanLinter runs over the
compiled freestyle ExecutionPlan (WorkflowGraph) before it is locked, complementing
ExecutionPlanCompiler (which throws on unreachable/unknown-kind/bad-edge) with the checks
it does NOT make:
- HARD unproduced_need — a stage needs an artifact no stage produces. Real gap: only the
TOML loader checked this; the freestyle compiler did not, so such plans compiled and
failed at runtime.
- HARD trap_state — a stage with no path to the terminal (inescapable loop / dead end).
Uses terminal-reachability, NOT "any cycle", so legitimate verdict-gated review loops pass.
- SOFT stage_count / fan_out / empty_brief / duplicate_brief — scored, never blocking.
FreestyleDriver lints after compile and records PlanLintCompletedEvent (pure function of
the recorded plan → replay-safe, no env observation in v1); a hard failure rejects the plan
(ExecutionPlanRejectedEvent source="lint") before the operator is asked or the plan locks,
so a broken plan never executes. Pays off now for single-plan freestyle; becomes the
per-candidate filter when multi-candidate generation (Slice 2) lands.
Deferred to later slices: file-path grounding + symbol resolution (needs an index),
token-budget/ADR-bypass checks, and a dedicated :core:planning module.
Run operator-configured static-analysis commands (compiler/detekt/formatters) as a
deterministic harness step on the producing stage, before the LLM reviewer. A stage
declares `static_analysis = [commands]`; after it produces its artifact, each command
runs in the workspace root and the results are recorded in a StaticAnalysisCompletedEvent
(invariant #9 env observation — recorded live, folded on replay, never re-run). A
non-clean command fails the stage retryably with its output fed back verbatim (§2:
compiler/test output is ground truth), so the implementer fixes it and only static-clean
code ever reaches the reviewer. This makes §5's "reviewer context excludes static findings"
mechanical — the findings are resolved upstream, so the reviewer can't waste inference on
what detekt catches for free; no output-parsing, no context plumbing.
- StaticAnalysisCompletedEvent + StaticAnalysisFinding (core:events) + registration.
- StaticAnalysisRunner seam + ProcessStaticAnalysisRunner (whitespace-split argv, stderr
merged, timeout→nonzero exit) in core:kernel; wired (nullable) into OrchestratorEngines
and the server. Null runner / no workspace root → logged no-op, never blocks the run.
- StageConfig.staticAnalysis + `static_analysis` TOML field; runStaticAnalysis folded into
a runPostStageGates sequence (produces → grounding → echo → static analysis).
- Documented `static_analysis` on the role_pipeline implementer stage (commented; commands
are workspace-specific). The reviewer prompt half shipped in 3467826.
Surface the health subsystem (llama/event-store/disk probes) in the TUI,
mirroring the session-stats pane: ClientMessage.GetHealthChecks ->
ServerMessage.HealthChecks (health.checks, reuses HealthReport) ->
StreamQueries.healthChecks via HealthInspectionService; Go OverlayHealth on
key H + palette 'health checks', pull-based fetch, per-subject status with
degraded loud. Empty/disabled -> visible fallback (no blank/lie). Go+Kotlin
golden tests pin the wire contract field-for-field. Observability spec §4/§3.
Third HealthProbe (after disk + llama): times a cheap lastGlobalSequence()
read as an event-store responsiveness heartbeat — DEGRADED on throw or when
latency exceeds eventStoreLatencyWarnMs (default 500). Read-only by design
(no probe writes into the source-of-truth log; Invariant #1). Plugs into the
existing HealthMonitor; no new events. Observability spec §4.
Also wraps a long line in LlamaServerHealthProbeTest to clear a detekt
MaxLineLength finding introduced in aaf896d.
Second HealthProbe (after disk-watermark): liveness via GET /health on the
configured llama base URL + tokens/sec degradation watch derived from recorded
InferenceCompletedEvents (windowed average vs configurable threshold, opt-in via
llamaTpsWarnBelow). Plugs into the existing HealthMonitor edge-detection loop;
no new events. Observability spec §4.
OverlayIdeas (opened with I, global/cross-session) lists the idea board in the
artifact-viewer shell: ↑↓ to move, x/d to remove (sends DiscardIdea, optimistic
drop), esc to close. Pull-based via ListIdeas -> idea.list.
ListIdeas -> idea.list (StreamQueries folds the board from the log, replay-neutral)
and DiscardIdea -> append IdeaDiscardedEvent landed in the capturing session, then
reply with the refreshed board. IdeaDto on the wire.
Render a workflow.proposed frame as a single-select picker with a manual-answer
slot: ↑↓/jk to choose, enter to launch, e for custom, esc to peek away. Picking
a candidate sends StartSession (and focuses the launched session); the custom
slot continues the conversation via ChatInput. Reuses the modal helpers from the
clarification view.
Map WorkflowProposedEvent to a workflow.proposed ServerMessage carrying the
candidates + original request. No new ClientMessage: a pick reuses StartSession
(launch the chosen workflow seeded with the request) and the manual-answer slot
reuses ChatInput.
Render stage clarification questions as an AskUserQuestion-style form:
header chip, prompt, selectable option chips, and a free-text custom slot.
Arrow/hjkl nav, space to select, e for custom, enter to submit. Submit
guards on all-answered and sends ClarificationResponse; the parked stage
re-runs with the answers in context.
Surfaces the clarification loop over the wire: ServerMessage.ClarificationRequired
(clarification.required, carrying the structured questions) is mapped from
ClarificationRequestedEvent; ClientMessage.ClarificationResponse carries the
operator's answers and is routed to orchestrator.submitClarification on both
the global and session streams. Reuses the core ClarificationQuestion/Answer
types as the wire shape. Round-trip tests for both directions.
Makes the research feature runnable end-to-end, off by default.
- config: [tools.research] (enabled, searxng_url, max_results, max_fetch_bytes).
- registration: web_search/web_fetch are built into BOTH the default and per-workspace
tool registries when research.enabled, sharing one HTTP client threaded from Main
(none built on the static path). Egress stays harness-enforced: web_fetch is T2
(operator-approved) and the existing NetworkHostRule still applies.
- workflow: examples/workflows/research.toml — decompose → gather → report, with the
three artifact schemas and prompts. Fan-out (search per sub-question, fetch per source)
runs as repeated tool calls inside the gather stage (Correx has no parallel agents);
per-source synthesis into the dossier is the compression step, so the report stage
consumes summaries, never raw pages. ResearchWorkflowTest validates the graph contract.
To run: set [tools.research].enabled, register the 3 [[artifacts]], copy research.toml +
prompts + schemas into the workflows dir, start SearXNG. Launch like any workflow (the
T2 fetch approval surfaces as an approval card; the report opens in the artifact viewer).
Follow-ups (noted, not blocking): batch fetch-approval at the source-list level (§3),
a dedicated SourceFetched/LowQualityExtraction event (quality + content hash are already
in tool-result metadata), dynamic per-session egress allowlist, and the web approval client (§6).
Health states are events. A HealthMonitor polls registered HealthProbes on an
interval and appends HealthDegraded/HealthRestored only on a status transition
(hysteresis), so health is observable, replayable, and renderable like every
other fact in the log. The probe reads nondeterministic environment state and
records it at observe-time (Invariant #9); HealthProjection rebuilds current
health by replay, never re-probing (Invariant #8). Monitor is live-only, started
in ServerModule.start alongside narration; seeded from the projection so a
restart does not re-emit a degraded the log already records.
First concrete probe: DiskWatermarkProbe — event-log + CAS size and growth rate,
warning before the 2am surprise. Read surface mirrors `correx stats`:
GET /health/checks replays the system session and `correx health` renders it.
Health events ride the reserved __system__ session, filtered from operator
session listings.
LLAMA_SERVER (liveness + tokens/sec trend) and EVENT_STORE (append/fsync latency)
probes plug into the same HealthProbe framework as follow-ups.
Scope creep — an implementer writing files it never declared — is the
dominant local-model failure that compiler/tests don't catch. A stage can
now declare a `writes` manifest (workspace-relative globs); a FILE_WRITE
that resolves inside the workspace but outside the declared set is raised
as PATH_OUTSIDE_MANIFEST → BLOCK, so the model gets the violation as tool
feedback and retries within scope (the §2 bounded-retry signal).
Implemented as a plane-2 rule beside PathContainmentRule (same effect-based
dispatch on FILE_WRITE, same symlink-safe WorldProbe resolution, same
recorded observations so replay reads them back — invariant #9). An empty
manifest is unrestricted; out-of-workspace targets stay PathContainmentRule's
concern, so the two rules don't double-flag.
- StageConfig.writeManifest + TomlWorkflowLoader `writes` parsing
- ToolCallAssessmentInput.writeManifest, threaded from the active stage via
SessionOrchestrator.runPlane2Assessment
- new ManifestContainmentRule, registered in the server assessor
- shared candidatePathStrings extracted so both path rules judge the same
target set
- role_pipeline.toml documents the option on the implementer stage
The dedicated ManifestViolationEvent the spec names is not added: the
violation is already recorded replayably in ToolCallAssessedEvent (issue +
observations + BLOCK) and surfaced in the TUI rationale band. A first-class
event is worth adding only once reconciliation consumes it.
Layer 1 of tui-requirements §5: a no-LLM analyzer that shell-lexes a
proposed command, classifies binaries against a known-command table, and
mechanically flags the constructs that widen blast radius — sudo/doas,
recursive delete, pipe-to-shell (skipping wrappers so `| sudo sh` counts),
redirects outside the workspace, network binaries, base64/eval, and
unparseable input (itself a red flag). Pure + replay-stable: same preview
+ workspaceRoot → same analysis, no environment reads.
The approval band renders command approvals as a card — worst-first flag
chips over the reconstructed command line, each token colored by severity
(T0–T4 ramp). ^x opens a scrollable fullscreen view so a long command is
never truncated. Diff and plain-preview paths are unchanged.
Server: computeToolPreview now renders the shell tool's argv as a full,
shell-quoted command line instead of the 200-char JSON-args fallback,
honouring §3's "full untruncated command". The TUI parser accepts both
the rendered line and the older `{"argv":[...]}` shape for replay.
Layer 2 (model annotation) deliberately deferred per the spec.
The TUI overflowed and became unreadable on narrow terminals: the status bar and
footer spilled far past the edge (justify never truncated), the two-column main
split shrank the events panel to ~20 cols, and the stats overlay wrapped mid-word.
- justify is now width-safe: shrinks the left segment first (the right holds the
high-signal connection clock / active spinner), clamps the right only as a last
resort, ANSI-aware so it never slices escape codes. No line can exceed the width.
- Main area collapses below narrowWidth (96): single full-width session list when
idle; output stacked over the live event strip in-session (the strip is too
valuable to drop — the constraint is width, not height).
- Compact chrome when slim: status drops workspace/gauge/provider-suffix and uses a
short clock ("◷ 3s") + bare spinner; footer uses a reduced hint set and drops the
"Soft·blue" badge; approval action row tightens its gaps so decision keys stay
visible.
- Overlays get near-full-width on slim terminals; stats pane uses compact formats +
per-line clipping so it never wraps.
Tests: no rendered line exceeds the terminal width at 40–160 cols (incl. the stats
overlay); narrow in-session main stacks output+events.
Implements the load-bearing TUI-requirements §2 hard requirements:
- Last-event clock: "last event Ns ago" always visible in the status bar while
in-session, updating every frame tick. Turns warn-colored when an active session
has gone quiet past the stale threshold — the "thinking vs hung" tell. The
WebSocket connection indicator (●/◌/○) already covers immediate drop visibility.
- Unknown-event raw fallback: applyServer now surfaces any unrecognized,
session-scoped event type as a raw row in the event stream instead of silently
dropping it (the frontend must not lie by omission). Recognized-but-unrendered
types (router.response, protocol_error) get an explicit no-op so the default
branch means genuinely unknown.
Tests: agoLabel formatting, unknown-event surfaces-as-row, known-ignored
no-spurious-row. Preview "session" frame exercises the clock.
Deferred: the full Go↔Kotlin per-event-type render matrix (§2 last bullet) — a
larger fixture effort than the core behaviors landed here.
Implements observability-spec §2 + §3-tier-1: metrics as a pure fold over the
event log, no third pillar, no OTel. Every metric is replayable over the full
historical log because it derives from events other contexts already emit — no
new EventPayload subclasses, so no serialization registry change.
- MetricsProjection: Projection<MetricsState> folding inference/tool/approval/
stage/workflow events into raw sums; approval request→decision latency is
correlated in-fold via a transient pending-request map.
- MetricsInspectionService: rebuilds via DefaultEventReplayer, derives read-side
ratios (token throughput, avg approval wait, session wall-time accounting split
into inference / tool / approval-wait / other) into MetricsReport.
- GET /sessions/{id}/metrics route, mirroring the replay endpoint.
- `correx stats <sessionId>` CLI client (--json passthrough), mirroring replay.
- Tests: seeded-event fold correctness with controlled timestamps (7) + CLI
render smoke tests (4).
kami