Commit graph

340 commits

Author SHA1 Message Date
Bryan Helmkamp
d0c8bd3eed Rename ExecutionEnvironment to Sandbox across the codebase
Adopt cleaner terminology: Sandbox (resource providing disk + execution),
SandboxProvider (Local/Docker/Daytona), SandboxEvent (lifecycle events),
and Snapshot (pre-built environment images). Flatten DaytonaSandboxConfig
into DaytonaConfig, rename CLI flag to --sandbox, and update TOML config
sections from [execution] to [sandbox].

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-02 11:33:21 -05:00
Bryan Helmkamp
ffa9d1d505 Rename docs site from Mint Starter Kit to Arc
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-02 11:21:33 -05:00
Bryan Helmkamp
e938d3395a Throw Response from apiJson so API 404s render as 404 pages
apiJson was throwing a plain Error on non-ok responses, which React
Router always treated as a 500. Now throws a Response with the actual
status code so API 404s surface as proper 404 error pages.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-02 11:05:34 -05:00
Bryan Helmkamp
fcba787cd5 Add styled 404 and 500 error pages matching Arc design system
Replace the unstyled error boundary with themed pages: teal gradient
number for 404, coral for 500, centered on the atmospheric background
with action buttons and a collapsible stack trace in dev mode.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-02 11:04:04 -05:00
Bryan Helmkamp
802727ac07 docs 2026-03-02 11:01:38 -05:00
Bryan Helmkamp
6fc2c1dcb0 Fill in all demo data gaps to match original React mock data
- Verifications: all 8 categories with 31 controls, all slugs return detail
  with 10 evaluations, 5 recent results, full checks, and sibling controls
- Run verifications: all 8 categories with correct pass/fail/na statuses
- Retros: per-run-ID retro detail for all 5 runs with full stages, learnings,
  friction points, open items, and correct field names
- Sessions: all 8 sessions in list, detail for s1/s2/s3, s1 expanded to 7
  turns with rich tool use data
- Runs: run-8 checks expanded to 12, run-9 to 6
- Settings: all 24 fields with full select options and descriptions
- Retro list files_touched expanded to full arrays

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-02 10:03:09 -05:00
Bryan Helmkamp
5c54f37a7c Add mintlify docs 2026-03-02 09:59:44 -05:00
Bryan Helmkamp
a456672959 move docs -> docs-internal 2026-03-02 09:59:00 -05:00
Bryan Helmkamp
4849dd8488 draft-api-design.md 2026-03-02 09:41:01 -05:00
Bryan Helmkamp
49364b0843 Add demo API server and wire React app to fetch from API
Expand the OpenAPI spec from 11 to 39 endpoints covering Runs, Workflows,
Verifications, Retros, Sessions, Insights, Settings, and Projects with ~45
schemas. Add `--demo` flag to `arc serve` that serves static demo data for
all endpoints (auth disabled, read-only). Non-demo mode returns 501 for new
endpoints while existing run handlers continue working.

Regenerate the TypeScript API client and add `apiJson` helper. Wire all 19
React route files with server-side loaders that fetch from the API and map
snake_case responses to camelCase UI types. Mock data kept as fallback.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-02 09:39:53 -05:00
Bryan Helmkamp
331dad98ec Rename Pipeline to Workflow in frontend
- routes/pipelines.tsx → routes/runs.tsx, component Pipelines → Runs
- Update route config to point to runs.tsx
- Retro.pipeline_name → workflow_name in data and all display components
- Table header "Pipeline" → "Workflow" in retros list

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-02 00:09:15 -05:00
Bryan Helmkamp
46af8e6887 Rename Pipeline to Workflow in CLI help text and test assertions
Update command descriptions and test event name strings to use
workflow/run terminology instead of pipeline.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-02 00:07:35 -05:00
Bryan Helmkamp
a566aa0352 Rename Pipeline to Run in arc-api server
- ManagedPipeline → ManagedRun, AppState.pipelines → AppState.runs
- All route paths: /pipelines/* → /runs/*
- Handler fns: list_pipelines → list_runs, start_pipeline → start_run,
  get_pipeline_status → get_run_status, cancel_pipeline → cancel_run
- Import renamed types: RunStatus, RunStatusResponse, StartRunRequest,
  StartRunResponse, WorkflowRunEngine, WorkflowRunEvent
- All test URIs, variable names, and test names updated

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-02 00:06:16 -05:00
Bryan Helmkamp
212de7a7ea Rename Pipeline to Workflow/Run in arc-workflows engine internals
- pipeline.rs → workflow.rs: PipelineBuilder → WorkflowBuilder,
  prepare_pipeline() → prepare_workflow()
- event.rs: PipelineEvent → WorkflowRunEvent, variant renames
  PipelineStarted/Completed/Failed → WorkflowRunStarted/Completed/Failed
- engine.rs: PipelineEngine → WorkflowRunEngine, pipeline_name → workflow_name
- retro.rs: pipeline_name → workflow_name, pipeline_failed → run_failed
- All handler, CLI, and test files updated to match

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 23:58:42 -05:00
Bryan Helmkamp
4c971d81b1 Rename Pipeline to Run in OpenAPI spec and generated types
Rename all API paths from /pipelines/* to /runs/*, schemas from
Pipeline* to Run*, and operation IDs accordingly. Regenerate both
the Rust types (arc-types) and TypeScript client (arc-api-client).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 23:48:07 -05:00
Bryan Helmkamp
7002f0638a Rename ManagerLoopHandler to SubWorkflowHandler
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 23:39:58 -05:00
Bryan Helmkamp
f45c1b1e2d Remove unused SubPipelineHandler
No DOT pipelines reference the sub_pipeline handler type. The manager
loop handler now covers the child pipeline spawning use case.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 23:38:53 -05:00
Bryan Helmkamp
47e16d4c0d Replace ChildObserver mock with real child engine spawning in ManagerLoopHandler
The manager loop (house node) now parses a child DOT pipeline, spawns a
real PipelineEngine, and monitors it via a tokio::select poll loop. Context
is cloned into the child and diffed on completion to propagate updates back
to the parent.

- Add PipelineEngine::from_services() to share parent's Arc services
- Add PipelineEngine::run_with_context() returning (Outcome, Context)
- Rewrite ManagerLoopHandler as unit struct with real child engine spawning
- Support both inline DOT (stack.child_dot_source) and file (stack.child_dotfile)
- Delete ChildObserver trait entirely
- Update all existing tests and add new e2e tests for context flow and dotfile

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 23:35:47 -05:00
Bryan Helmkamp
1aca8231bc Add Terminal dropdown button to run detail page
Adds a sandboxId field to RunItem and a Terminal dropdown on the run
detail page with "Web Terminal" (links to Daytona proxy) and
"Connect with SSH" options. Uses existing Headless UI Menu pattern.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 23:01:15 -05:00
Bryan Helmkamp
726ab184dc Add cursor: pointer to all click targets site-wide
Add a global CSS base rule for interactive elements (buttons, links,
selects, summary, ARIA roles) and add cursor-pointer to clickable
table rows. Remove now-redundant per-element cursor-pointer classes.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 22:36:21 -05:00
Bryan Helmkamp
f6bc346663 Fix remote parallel worktree path to mirror host structure
Replace hardcoded /home/daytona/workspace/.arc-parallel path with
{working_directory}/.arc/logs/{run_id}/parallel/{node_id}/{branch_key},
matching the host worktree layout and scoping worktrees per-run.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 22:27:56 -05:00
Bryan Helmkamp
ae72cd059d Add verification control detail page with clickable catalog rows
Adds /verifications/:slug route with detail page showing control
description, stat cards, evaluation history, recent runs, checks/examples,
and sibling navigation. Makes verification catalog rows fully clickable.
Fixes dark mode hydration mismatch with suppressHydrationWarning.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 22:25:00 -05:00
Bryan Helmkamp
bda63295ed Fix model test: increase timeout to 30s and min tokens to 16
OpenAI requires max_output_tokens >= 16, and slower providers like
Gemini Pro need more than 10s to respond.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 22:14:50 -05:00
Bryan Helmkamp
4b3a450071 Fix CLI integration tests and clippy warnings
- Update CLI tests for restructured commands (models → top-level, run → run start)
- Remove sync-related tests (command was removed)
- Fix clone_on_copy clippy warnings in arc-api server

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 22:07:12 -05:00
Bryan Helmkamp
b2b64dcc64 Restructure CLI: nest runs under arc run, promote models to top-level
- `arc run <pipeline>` → `arc run start <pipeline>`
- `arc runs list/prune` → `arc run list/prune`
- `arc llm models list` → `arc models list`
- Remove `arc llm models sync`
- Add `arc models test [--provider, --model]` for connectivity testing
- Clean up dead code (RunsArgs, RunsCommand, runs_command)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 22:04:32 -05:00
Bryan Helmkamp
4a0fc63d5a Add generated Axios API client package (@qltysh/arc-api-client)
Replace openapi-typescript types-only output with a full TypeScript Axios
client generated by openapi-generator-cli. The generated code is committed
so IDE support works without running codegen. arc-web consumes the client
via bun workspaces.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 22:03:38 -05:00
Bryan Helmkamp
1334a952af Add spec-first OpenAPI for arc-api with generated types
- Create openapi/arc-api.yaml as source of truth for all API endpoints
- Add arc-types crate with build.rs using typify to generate Rust structs
  from the spec's component schemas
- Refactor server.rs to use generated types instead of hand-written ones
- Add route coverage conformance test validating router matches spec
- Add openapi-typescript to arc-web for TypeScript type generation

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 21:43:49 -05:00
Bryan Helmkamp
95864ab788 Add run labels and arc runs list/prune CLI
- Add `labels: HashMap<String, String>` to RunConfig, written to manifest.json
- Add `--label KEY=VALUE` flag to `arc run` (repeatable)
- New `arc runs` command: list pipeline runs with table or --json output
- New `arc runs prune` command: delete old runs with --before, --pipeline,
  --label, --orphans filters (dry-run by default, --yes to confirm)
- 13 new tests covering scan_runs, filter_runs, prune, and manifest labels

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 20:23:26 -05:00
Bryan Helmkamp
c860bc3e35 Suppress clippy::too_many_arguments on derive_retro
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 19:47:08 -05:00
Bryan Helmkamp
d7d2294b7c Add structured tracing across 7 crates per logging audit plan
Implements all 47 gaps and fixes all 7 violations identified in
docs/agent/logging-audit-plan.md:

- arc-llm: Add tracing dep + 18 log statements (requests, responses,
  retries, timeouts, tool execution)
- arc-api: Add tracing dep + 6 log statements (server lifecycle,
  pipeline start/complete/fail), replace eprintln with warn
- arc-mcp: Fix 3 violations (string interpolation, secret leakage),
  add 7 log statements (client creation, handshake, tool calls)
- arc-workflows: Fix 3 eprintln violations, add 9 log statements
  (git checkpoints, Daytona sandbox, worktrees, node visit limit)
- arc-git-storage: Add tracing dep + 9 log statements (branch ops,
  snapshot write/delete/rename)
- arc-db: Add 3 log statements (connection, migrations)
- arc-agent: Add 4 debug statements (session init, compaction)
- arc-cli: Add 1 debug statement (command dispatch)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 19:43:19 -05:00
Bryan Helmkamp
b15a526b0a Add darker logotype variant for light mode
The existing gradient logo (#c4dede-#495669) was designed for dark
backgrounds. Adds logotype-light.svg with the same paths but a
deeper gradient (#2a6b6b-#1c2a3a) for better contrast on white,
and swaps the src based on the current theme.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 19:37:15 -05:00
Bryan Helmkamp
38da286130 Add stall watchdog to detect hung pipeline handlers
A background tokio task polls EventEmitter.last_event_at(). When idle
time exceeds the graph-level stall_timeout (default 600s), it cancels
a CancellationToken that races against execute_with_retry via
tokio::select!, dropping the hung handler future.

- EventEmitter: AtomicI64 last_event_at field, touch() to seed, emit()
  auto-updates
- Graph::stall_timeout(): reads Duration attr, defaults 600s, None for 0
- StallWatchdogTimeout event variant with warn-level trace()
- Engine: watchdog spawn before main loop, select! at handler call,
  shutdown after loop
- CLI: format arms for summary and detail views
- Unit tests for emitter, graph accessor, event serialization, and
  engine watchdog behavior (hung, keepalive, disabled)
- E2e integration tests: DOT-parsed pipelines with 100-200ms stall
  timeouts verifying trigger, keepalive, disable, and timing

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 19:36:58 -05:00
Bryan Helmkamp
e515368838 Link ARC logo to /start instead of /workflows
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 19:33:02 -05:00
Bryan Helmkamp
e00898d266 Document event enum tracing and double-logging prevention
Update logging strategy to explain that event enums have trace()
methods called automatically at emit sites, and that wrapper
variants are no-ops to prevent double-logging.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 19:18:01 -05:00
Bryan Helmkamp
59f4576f8f Fix SubAgentEvent double-logging by making trace() a no-op
Inner events are already traced at origin by the sub-agent's own
EventEmitter, so the SubAgentEvent wrapper should not re-trace them.
This matches the pattern used by PipelineEvent::Agent and
PipelineEvent::ExecutionEnv.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 19:17:18 -05:00
Bryan Helmkamp
ec70be1e0d Add trace() method to event enums for structured file logging
Every emitted event now produces a structured tracing log line so
developers can debug after the fact via ~/.arc/logs/. Each event
variant gets an appropriate log level (info/debug/warn/error) with
structured fields. Streaming noise variants (TextDelta,
ToolCallOutputDelta) are no-ops, and wrapper variants (Agent,
ExecutionEnv on PipelineEvent) delegate to the inner event's trace.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 19:15:43 -05:00
Bryan Helmkamp
5ea508a5ca Add stable failure signature hints for model errors
SdkError now produces hand-crafted signature hints (e.g.
"api_deterministic|openai|authentication") that are identical regardless
of error message wording, replacing fragile regex-normalized signatures
for API errors. ArcError.to_fail_outcome() centralizes fail outcome
construction with failure_class and failure_signature context_updates.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 19:12:29 -05:00
Bryan Helmkamp
3b8a0c3af2 Fix light mode colors for diff views and charts
The .light CSS variable overrides were defined before the :root dark
defaults, so equal-specificity source-order caused the dark values to
always win. Move .light overrides after each :root block so they
correctly take effect.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 19:12:13 -05:00
Bryan Helmkamp
e87ca6a8af Block non-transient_infra failures from loop_restart and fix UTF-8 truncation
Guard loop_restart edges to only allow transient_infra failures, matching
Kilroy's behavior. Non-transient failures (deterministic, structural,
budget_exhausted, canceled, compilation_loop) are now blocked immediately
instead of getting restart attempts before the circuit breaker fires.

Fix normalize_failure_reason truncation to use floor_char_boundary(240)
instead of a raw byte slice, preventing panics on multi-byte UTF-8 chars.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 19:00:39 -05:00
Bryan Helmkamp
b18641277b Align max_node_visits to use >= like circuit breakers, matching Kilroy
max_node_visits used > while signature circuit breakers used >=, creating
an inconsistency where "limit" meant different things depending on the
mechanism. Kilroy uses >= for all three checks. This aligns Arc to match.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 18:29:26 -05:00
Bryan Helmkamp
303c78e584 Add logging strategy document
Defines structured logging conventions for the team: level guidelines
(INFO for production, DEBUG for on-demand investigation), structured
field patterns, and per-crate examples.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 18:29:14 -05:00
Bryan Helmkamp
716400ac06 Add file-based tracing infrastructure with ARC_LOG control
Tracing events (e.g. 13 calls in arc-mcp) were silently dropped because
no subscriber was configured. This adds a file-based tracing subscriber
that logs to ~/.arc/logs/YYYY-MM-DD.log with INFO as the default level,
controllable via the ARC_LOG env var.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 18:18:12 -05:00
Bryan Helmkamp
1a95cd6664 Rename AgentBackend → AgentApiBackend and CliBackend → AgentCliBackend
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 18:14:25 -05:00
Bryan Helmkamp
a535c6575f Rename kilroy → attractor in test directory and compat tests
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 18:10:22 -05:00
Bryan Helmkamp
3c7b786ed1 Apply cargo fmt formatting
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 18:08:49 -05:00
Bryan Helmkamp
2ed164d190 Add light mode support with theme toggle
Introduce semantic CSS custom properties (page, panel, fg, line, overlay,
etc.) that swap values via .light/.dark class on <html>. An inline script
reads localStorage / prefers-color-scheme before first paint to prevent
flash-of-wrong-theme.

- ThemeProvider + useTheme hook in app/lib/theme.tsx
- Sun/Moon toggle in desktop nav and mobile menu
- .light overrides for all semantic tokens, accent colors (WCAG AA on
  white), atmosphere gradient, @pierre/diffs surfaces, and chart variables
- Migrated ~30 files from hardcoded color classes to semantic tokens
- GraphViz diagrams use getGraphTheme() for light/dark hex maps
- Chart gridlines and axis labels use CSS custom properties
- Workflow card colors reference CSS vars for automatic theme switching
- Pierre diffs switch between pierre-dark and pierre-light themes

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 18:06:45 -05:00
Bryan Helmkamp
cab25a3e7e Add failure signatures and circuit breakers for deterministic failure cycle detection
Introduces reason normalization (strip variable data like line numbers,
hex hashes), composite failure signatures (node_id|class|normalized_reason),
and circuit breakers that track signature counts to abort when the same
deterministic failure repeats beyond a configurable limit (default 3).

Key additions:
- normalize_failure_reason() strips hex, digits, whitespace for stable grouping
- FailureSignature type with handler-provided hint priority
- FailureClass::is_signature_tracked() (deterministic + structural only)
- Graph-level loop_restart_signature_limit attribute
- LoopState struct bundling node_visits + signature maps through run_internal
- Loop failure circuit breaker (same node repeating)
- Restart failure circuit breaker (across loop_restart edges)
- Checkpoint persistence for both signature maps with backward compat
- 18 e2e integration tests covering all circuit breaker scenarios

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 18:03:09 -05:00
Bryan Helmkamp
f53cc746dc Add split/unified and background toggles to diff headers
Inject per-file toggle controls into the Pierre diff header bar via
renderHeaderMetadata. Each file gets independent split/unified and
background on/off state using the Pierre icon style.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 17:43:57 -05:00
Bryan Helmkamp
24b454b145 Fix retro error handling and add Session wall-clock timeout
- Add wall_clock_timeout field to SessionConfig that spawns a tokio timer
  to cancel the session after a duration, reusing existing Aborted path
- Set 120s wall-clock timeout on retro agent to prevent unbounded runs
- Replace silent `let _ =` with logged warnings for checkpoint load and
  retro save failures
- Reuse LLM client from initial from_env() call instead of creating a
  second one for the retro agent
- Extract retro generation into generate_retro() helper and call it from
  both run_command and run_from_branch (resume path)
- Tolerate mutex poisoning in retro agent with unwrap_or_else

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 17:39:13 -05:00
Bryan Helmkamp
375a833733 Add Steer button and inline comment forms to diff view
Hover any line in a diff hunk to see a "Steer" button on the right side.
Clicking opens an inline comment form; submitted comments persist as
read-only annotations. Uses onLineEnter/onLineLeave with absolute overlay
positioning to place the button at the row's right edge.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-01 17:16:27 -05:00