autoresearch-run

name	autoresearch-run
description	Use when running the autoresearch loop for a project whose survey and validator gates have passed — executes batches of attempts (one git worktree + LOG.md each, scored only by the validator under a hard time limit), then reflects, reports, and re-plans; continues autonomously while authorized_rounds remain, otherwise stops for user review. Stage 4 of the autoresearch pipeline.

Autoresearch Run

The loop. Protocol per attempt: references/attempt-protocol.md. Report per cycle: references/reflection-template.md. Configuration from research/STATE.md: batch_size (default 10), time_limit_seconds, authorized_rounds, next_attempt, next_cycle.

Entry check — refuse to start otherwise

Both survey_gate and validator_gate must read passed in research/STATE.md, and their artifacts must verify on disk (CATALOG.md, INSIGHTS.md ## Selected, validator manifest with self-test results). If not, refuse and route back through the autoresearch dispatcher. A missing gate is never worked around; a user-approved exception goes into overrides: in STATE.md first.

Hard rules (non-negotiable during the loop)

1. Every attempt in its own .worktrees/attempt-NNN/ with a LOG.md.
2. Every scored run goes through the validator CLI; nothing else counts.
3. Hard wall-clock limit time_limit_seconds on every scored run.
4. Holdout labels never enter design context — only aggregate validator output.
5. Crashes/timeouts are recorded failures, never silently retried.

Cycle

1. Plan the batch. Generate a surplus of candidate hypotheses (~2× batch_size), each drawing on ## Selected entries in research/INSIGHTS.md, then rank them with a quick rubric (expected information gain, cost, distinctness) and promote the top batch_size. Two filters before anything is implemented:
   • Novelty check — compare each candidate against the hypotheses in all prior attempts' LOG.md files; a near-duplicate of anything already tried is rejected and resampled. Never re-spend an attempt on a restated old idea.
   • Batch composition — mix drafts (genuinely different approaches) with improvements (exactly one atomic change to the best-scoring known-good ancestor, so the change's effect is measurable) and at most 2 debug attempts on a promising-but-broken branch. A failing branch that exhausts its debug cap is abandoned, not nursed. Scope the planning context: drafts see a digest of sibling attempts (what was tried, what it scored — do something different); improvements and debugs see their ancestral chain (avoid undo-redo loops). Feed forward the failure artifacts (validator errors[], stderr) of the previous batch — failures are data.
2. Execute each attempt per references/attempt-protocol.md.
3. Reflect. Write docs/discussion/YYYY-MM-DD-HHMMSS-cycle-NN.md per references/reflection-template.md; increment next_cycle. Dev-score selection overfits over long runs: if the holdout query budget in the validator manifest allows, adjudicate the cycle's top candidate on the holdout (aggregate result only) and record it.
4. Soft gate. Decrement authorized_rounds:
   • if > 0 remain: continue to the next cycle autonomously;
   • if 0: stop, present the report and the proposed next-batch plan, and wait for the user to re-authorize (set a new authorized_rounds), amend the plan, or stop. Insight promotions proposed in the report are confirmed here.

Termination

When the validator reports the GOAL.md bar met on dev instances, run validate --instances holdout once, report both results, set stage: done, and hand off to the user — write-up is out of scope for this skill (use paper-writer).