Mission Control

Mission Control is HAR’s open-source local dashboard. It combines run records and slot state from registered repositories without replacing the repository-owned harness.

Start the packaged dashboard

har control up

The CLI pulls the Mission Control image matching its own version and starts the app plus PostgreSQL with Docker Compose. Build the image from a source checkout instead:

har control up --build

Stop it with har control down.

Register and sync

har env init remembers a repository for synchronization. You can also register one explicitly:

har control register --repo /path/to/project
har control sync --repo /path/to/project
har control watch --interval 10

watch can sync one repository or all registered repositories continuously. --dry-run previews registration or sync, and sync --json produces structured output.

What the dashboard shows

The repository overview summarizes registered projects, total runs, active slots, and harness profiles. Repository detail provides:

• active slots, worktrees, branches, preview URLs, dirtiness, and drift;
• a run timeline with stage, trigger, duration, and status;
• expected validation stages and their latest results;
• exact-tree change batches and associated commits;
• files under .har/artifacts/;
• verification trends and pass rates.

The global Worktrees home page (/) highlights dirty, stale, detached, and bypass-warning sessions across repositories. It also shows token and cost columns when agent telemetry is enabled (see below). Click a slot id for the slot leaf (LLM usage + Verify pipeline + session timeline). Repo catalog lives at /repos; cross-repo usage rollup at /usage.

Agent usage telemetry (Claude Code / Codex)

HAR can attribute Claude Code and Codex token usage (and Claude cost estimates) to each worktree/session and show it in Mission Control. Preference shape:

{ "enabled": true, "signals": { "metrics": true, "logs": true, "prompts": false, "traces": false } }

Defaults when telemetry is on: metrics + logs (events without prompt bodies). Prompt text is opt-in.

har telemetry status
har telemetry on              # metrics + logs; ensure Mission Control is running
har telemetry on --prompts    # also ship user/assistant text to local MC
har telemetry on --traces     # thin traces + Claude enhanced telemetry beta
har telemetry off             # stop OTEL injection and usage harvest; keeps historical rows

Preference is stored in ~/.har/telemetry.json. Override with HAR_TELEMETRY=0|1.

When telemetry is on:

1. har env launch auto-starts Mission Control if it is not reachable (har control up).
2. Launch writes session tags and OTEL exporter env into .env.agent.<id> (Claude Code), including OTEL_LOGS_EXPORTER=otlp when logs are enabled.
3. Agents export metrics to {HAR_CONTROL_API_URL}/api/otel/v1/metrics and logs to /api/otel/v1/logs (traces to /api/otel/v1/traces when opted in).
4. har control sync also harvests local Claude/Codex session files as a fallback (prompt summaries only when signals.prompts is true).

Privacy: prompt/response text leaves the agent machine only when the prompts signal is on. Mission Control stores usage and events in local SQLite.

For Codex, print a config snippet (merge into ~/.codex/config.toml yourself):

har telemetry codex-snippet --agent-id 1 --write

Disable anytime: har telemetry off.

Local development

Mission Control’s source has its own harness:

cd control
har env launch 1

Use this path for development, manual testing, and screenshots. Do not run it on the same port as the packaged har control up instance. Harness preflight detects the conflict and can select another port in the slot lane.

HAR Cloud

har control login --api-key ... configures a hosted API key for the current process, and har control sync --cloud targets HAR Cloud. Hosted coordination is separate from the local open-source dashboard and portable .har/ contract.