autoresearch

Files (1)

• SKILL.md

SKILL.md (first lines)

# Autoresearch

Autoresearch is the skill-first validator-gated research loop for Omagy. It is
the Omagy adaptation of OMX `$autoresearch`: keep the measured research loop,
but run it as an Antigravity skill with Omagy state and artifact gates instead
of a direct detached CLI launch.

## Use When

- You want a Ralph-ish persistent research loop.
- The task should keep nudging until explicit validation evidence exists.
- You want init-time choice between script validation and prompt+architect
  validation.

## Do Not Use When

- You want the old direct `omx autoresearch` command surface.
- You want detached tmux or split-pane launch parity.
- You have not decided the validation regime yet.

## Core Contract

1. **Init chooses validation mode.** Pick exactly one:
   - `mission-validator-script`
   - `prompt-architect-artifact`
2. **Persist mode state** in `.omagy/state/autoresearch-state.json` including:
   - `validation_mode`
   - `completion_artifact_path`
   - `mission_validator_command` or `validator_prompt`
   - optional `output_artifact_path`
3. **Completion is artifact-gated.** The loop does not stop because the model
   says "done", because a stop hook fired once, or because several turns were
   no-ops.
4. **Direct launch is intentionally unsupported.** Use
   `/omagy:deep-interview --autoresearch` for intake and `/omagy:autoresearch`
   for execution.

## Completion Artifact Contract

### `mission-validator-script`

The completion artifact must exist and record a passing validator result:

```json
{
  "status": "passed",
  "passed": true,
  "summary": "metric improved beyond baseline"
}
```

### `prompt-architect-artifact`

The completion artifact must include both an architect approval verdict and an
output artifact path:

```json
{
  "validator_prompt": "Review the research output against the mission.",
  "architect_review": { "verdict": "approved" },
  "output_artifact_path": ".omagy/specs/autoresearch-demo/report.md"
}
```

## Recommended Flow

1. Run `/omagy:deep-interview --autoresearch` to clarify mission + evaluator.
2. Materialize `.omagy/specs/autoresearch-<slug>/mission.md`, `sandbox.md`,
   and `result.json`.
3. Start `/omagy:autoresearch` with the chosen validation mode:

```bash
omagy autoresearch start --input '{"task":"<mission>","validation_mode":"mission-validator-script","completion_artifact_path":".omagy/specs/autoresearch-demo/result.json","mission_validator_command":"<validator command>"}' --json
```

4. Continue research until the completion artifact satisfies the chosen mode.
5. Finish only after the validator artifact is complete:

```bash
omagy autoresearch complete --json
```

## Migration Note

- No direct CLI launch.
- No detached tmux split-pane launch.
- No noop-count completion gate.

From shayne-snap/oh-my-antigravity, MIT.

More from shayne-snap/oh-my-antigravity