har

Read manifest, stack hints, available scripts, and harness stages for a repository.

Scaffold .har/ boilerplate for a coding agent to adapt. Docker is required (Mission Control container + harness infra); the response reports Docker availability.

Validate .har/ against the bundled templates: returns validation issues, drift, and a maintenance bundle report. Pass finalize=true (optionally with summary) to record a completed manual adaptation in .har/manifest.json.

Install a verification plugin (bundled id, local path, npm package, or git URL) that registers stages in .har/stages.json.

Start a FRESH agent session from the main checkout HEAD (switch that checkout to main first for a new unrelated task). Run BEFORE editing any file: returns workDir — ALL edits go there, never the main checkout. Occupied slots always block: call har_get_status, then har_complete_environment or har_te…

Append a related external link (PR, mirrored issue, alternate tracker) to an existing work unit.

Resume a failed or partial agent launch without replacing the worktree. Alias for har_launch_environment with resume=true.

Readiness gate before launch: checks ports, foreign PM2, Docker conflicts, occupied slot, and untracked paths that will be missing from a session worktree. Returns canLaunch with actionable blockers and warnings. Call before har_launch_environment.

Run one generic harness stage by id or kind.

Run the project verification pipeline for an agent slot. Returns status, timing, and failed-step output (passing steps omit logs; stdout is not the raw JSON dump).

Validate the harness contract: harness.env schema, stages.json, stage scripts exist and are executable, lifecycle stages resolve, verificationStages ids resolve, port lanes are coherent, slot registry entries point at existing worktrees. Returns pass/fail with actionable findings.

Return structured slot status for one agent or all slots (same source as har env status/--json). Call BEFORE har_launch_environment when a slot may already be in use — shows worktree path, dirty state, branch, and readiness.

Return recent logs for a slot/process.

Stop a running agent environment slot. The session git branch is kept unless deleteBranch=true.

Finish a session when the work is done: reuses the last passing full validation for the current worktree (or re-runs verify when verify=true), tears the slot down, and KEEPS the session branch so the user can push it and open a PR.

List result JSON, screenshots, traces, reports, or other files under .har/artifacts/.

List persisted harness run records from .har/runs/.

Fetch one harness run record by runId.

Start local Mission Control (a single self-contained Docker container backed by SQLite) and sync all harness repositories that were scaffolded with har onboard or har_init_harness.

Scaffold a project-owned factory line at .har/lines/<id>/ (manifest, program, optional gate stage, README). A line is a multi-station program, not a verification plugin.

Install a factory line bundle (local id, path, npm package, or git URL). Registers the line's stages but NEVER adds them to verificationStages — default verify stays unchanged. Verification plugins use har_add_plugin instead.

Stations, cumulative gate progress derived from .har/runs/ records, and slots in flight for installed factory lines. Pure read — writes no run records.

Run one station's cumulative gate: every gate stage tagged at that station or earlier. Runs through the normal stage runner and writes run records; does not call verify and does not widen the verify plan.