auto-review-loop

> Override for Codex users who want Gemini, not a second Codex agent, to act as the reviewer. Install this package after skills/skills-codex/*.

Auto Review Loop: Autonomous Research Improvement

> Gemini overlay assurance: review_independence: cross-family and acceptance_status: accepted.

Autonomously iterate: review → implement fixes → re-review, until the external reviewer gives a positive assessment or MAX_ROUNDS is reached.

Context: $ARGUMENTS

Constants

• MAX_ROUNDS = 4
• POSITIVE_THRESHOLD: score >= 6/10 AND verdict ∈ {"ready", "almost"} — both must hold, matching the operative STOP CONDITION below. Verdict vocabulary is {"ready", "almost", "not ready"}. (Earlier wording used "or" + a stale verdict set; the AND form is authoritative.)
• REVIEW_DOC: review-stage/AUTO_REVIEW.md (cumulative log) (fall back to ./AUTO_REVIEW.md for legacy projects)
• OUTPUT_DIR = review-stage/ — Directory for review output files.
• REVIEWER_MODEL = gemini-review — Gemini reviewer invoked through the local gemini-review MCP bridge. Set GEMINI_REVIEW_MODEL if you need a specific Gemini model override.
• HUMAN_CHECKPOINT = false — When true, pause after each round's review (Phase B) and present the score + weaknesses to the user. Wait for user input before proceeding to Phase C. The user can: approve the suggested fixes, provide custom modification instructions, skip specific fixes, or stop the loop early. When false (default), the loop runs fully autonomously.

> 💡 Override: /auto-review-loop "topic" — human checkpoint: true

State Persistence (Compact Recovery)

Long-running loops may hit the context window limit, triggering automatic compaction. To survive this, persist state to review-stage/REVIEW_STATE.json after each round:

{
  "round": 2,
  "thread_id": "019cd392-...",
  "status": "in_progress",
  "last_score": 5.0,
  "last_verdict": "not ready",
  "pending_experiments": ["screen_name_1"],
  "timestamp": "2026-03-13T21:00:00"
}

Write this file at the end of every Phase E (after documenting the round). Overwrite each time — only the latest state matters.

On completion (positive assessment or max rounds), set "status": "completed" so future invocations don't accidentally resume a finished loop.

Workflow

Initialization

1. Check for review-stage/REVIEW_STATE.json (fall back to ./REVIEW_STATE.json if not found — legacy path):

• If neither path exists: fresh start (normal case, identical to behavior before this feature existed)
• If it exists AND status is "completed": fresh start (previous loop finished normally)
• If it exists AND status is "in_progress" AND timestamp is older than 24 hours: fresh start (stale state from a killed/abandoned run — delete the file and start over)
• If it exists AND status is "in_progress" AND timestamp is within 24 hours: resume
• Read the state file to recover round, thread_id, last_score, pending_experiments
• Read review-stage/AUTO_REVIEW.md to restore full context of prior rounds (fall back to ./AUTO_REVIEW.md)
• If pending_experiments is non-empty, check if they have completed (e.g., check screen sessions)
• Resume from the next round (round = saved round + 1)
• Log: "Recovered from context compaction. Resuming at Round N."

1. Read project narrative documents, memory files, and any prior review documents
2. Read recent experiment results (check output directories, logs)
3. Identify current weaknesses and open TODOs from prior reviews
4. Initialize round counter = 1 (unless recovered from state file)
5. Create/update review-stage/AUTO_REVIEW.md with header and timestamp

Loop (repeat up to MAX_ROUNDS)

Phase A: Review

Send comprehensive context to the external reviewer:

mcp__gemini-review__review_start:
  prompt: |
    [Round N/MAX_ROUNDS of autonomous review loop]

    [Full research context: claims, methods, results, known weaknesses]
    [Changes since last round, if any]

    Please act as a senior ML reviewer (NeurIPS/ICML level).

    1. Score this work 1-10 for a top venue
    2. List remaining critical weaknesses (ranked by severity)
    3. For each weakness, specify the MINIMUM fix (experiment, analysis, or reframing)
    4. State clearly: is this READY for submission? Yes/No/Almost

    Be brutally honest. If the work is ready, say so clearly.

    === SCOPE LIMITS (these bound what you PROPOSE, never what you look for) ===
    Report anything that is actually wrong here — including a rare-looking case, if
    this repo actually produces it. Then keep the fix in scope:
    1. This is a RESEARCH-WORKFLOW tool, not a security paper. Verification is
       welcome; over-defense is not. Assume a cooperating operator on their own
       machine — a malicious local user is NOT in the threat model.
    2. Do NOT propose SHA / hash / content-fingerprint / digest-binding schemes.
       Reporting a real defect in hashing code that already exists is fine.
    3. NO speculative machinery: do not add feature flags, migration frameworks,
       compat layers, wrappers, pins, or similar mechanisms unless evidence shows
       a current repo defect they fix or an explicit existing invariant they must
       preserve. "Load-bearing", "compatibility", and "not scaffolding" are labels,
       not evidence. Point to the failing path/artifact or invariant, and check the
       proposal's factual premises, such as whether a named package version exists.
    4. NO corner-case obsession: exotic encodings, symlink races, RTL text and
       millisecond races are out of scope unless you can show the case arises here.
    5. Where a rubric or checklist is genuinely needed, do not over-mechanize
       judgement. A clear sentence a human reads beats a scored table nobody
       maintains.
    Exception: code that runs remote commands, starts a network service, or installs
    an MCP server runs on the user's machine with their credentials — trust-boundary
    findings there are in scope and the default is strict.
    Say plainly when something is correct. Do not manufacture findings.

After this start call, immediately save the returned jobId and poll mcp__gemini-review__review_status with a bounded waitSeconds until done=true. Treat the completed status payload's response as the reviewer output, and save the completed threadId for any follow-up round.

If this is round 2+, use mcp__gemini-review__review_reply_start with the saved completed threadId, then poll mcp__gemini-review__review_status with the returned jobId until done=true to maintain continuity.

Phase B: Parse Assessment

CRITICAL: Save the FULL raw response from the external reviewer verbatim (store in a variable for Phase E). Do NOT discard or summarize — the raw text is the primary record.

Then extract structured fields:

• Score (numeric 1-10)
• Verdict ("ready" / "almost" / "not ready")
• Action items (ranked list of fixes)

STOP CONDITION: If score >= 6 AND verdict ∈ {"ready", "almost"} (exact match — "not ready" does NOT qualify) → stop loop, document final state.

Human Checkpoint (if enabled)

Skip this step entirely if HUMAN_CHECKPOINT = false.

When HUMAN_CHECKPOINT = true, present the review results and wait for user input:

📋 Round N/MAX_ROUNDS review complete.

Score: X/10 — [verdict]
Top weaknesses:
1. [weakness 1]
2. [weakness 2]
3. [weakness 3]

Suggested fixes:
1. [fix 1]
2. [fix 2]
3. [fix 3]

Options:
- Reply "go" or "continue" → implement all suggested fixes
- Reply with custom instructions → implement your modifications instead
- Reply "skip 2" → skip fix #2, implement the rest
- Reply "stop" → end the loop, document current state

Wait for the user's response. Parse their input:

• Approval ("go", "continue", "ok", "proceed"): proceed to Phase C with all suggested fixes
• Custom instructions (any other text): treat as additional/replacement guidance for Phase C. Merge with reviewer suggestions where appropriate
• Skip specific fixes ("skip 1,3"): remove those fixes from the action list
• Stop ("stop", "enough", "done"): terminate the loop, jump to Termination

Feishu Notification (if configured)

After parsing the score, check if ~/.codex/feishu.json exists and mode is not "off":

• Send a review_scored notification: "Round N: X/10 — [verdict]" with top 3 weaknesses
• If interactive mode and verdict is "almost": send as checkpoint, wait for user reply on whether to continue or stop
• If config absent or mode off: skip entirely (no-op)

Phase C: Implement Fixes (if not stopping)

For each action item (highest priority first):

1. Code changes: Write/modify experiment scripts, model code, analysis scripts
2. Run experiments: Deploy to GPU server via SSH + screen/tmux
3. Analysis: Run evaluation, collect results, update figures/tables
4. Documentation: Update project notes and review document

Prioritization rules:

• Skip fixes requiring excessive compute (flag for manual follow-up)
• Skip fixes requiring external data/models not available
• Prefer reframing/analysis over new experiments when both address the concern
• Always implement metric additions (cheap, high impact)

Phase D: Wait for Results

If experiments were launched:

• Monitor remote sessions for completion
• Collect results from output files and logs

Phase E: Document Round

Append to review-stage/AUTO_REVIEW.md:

## Round N (timestamp)

### Assessment (Summary)
- Score: X/10
- Verdict: [ready/almost/not ready]
- Key criticisms: [bullet list]

### Reviewer Raw Response

<details>
<summary>Click to expand full reviewer response</summary>

[Paste the COMPLETE raw response from the external reviewer here — verbatim, unedited.
This is the authoritative record. Do NOT truncate or paraphrase.]

</details>

### Actions Taken
- [what was implemented/changed]

### Results
- [experiment outcomes, if any]

### Status
- [continuing to round N+1 / stopping]

Write review-stage/REVIEW_STATE.json with current round, agent id, score, verdict, and any pending experiments.

Increment round counter → back to Phase A.

Termination

When loop ends (positive assessment or max rounds):

1. Update review-stage/REVIEW_STATE.json with "status": "completed"
2. Write final summary to review-stage/AUTO_REVIEW.md
3. Update project notes with conclusions
4. If stopped at max rounds without positive assessment:

• List remaining blockers
• Estimate effort needed for each
• Suggest whether to continue manually or pivot

1. Feishu notification (if configured): Send pipeline_done with final score progression table

Output Protocols

> Follow these shared protocols for all output files:

> - [Output Versioning Protocol](../../shared-references/output-versioning.md) — write timestamped file first, then copy to fixed name

> - [Output Manifest Protocol](../../shared-references/output-manifest.md) — log every output to MANIFEST.md

> - [Output Language Protocol](../../shared-references/output-language.md) — respect the project's language setting

Key Rules

• Large file handling: If the Write tool fails due to file size, immediately retry using Bash (cat << 'EOF' > file) to write in chunks. Do NOT ask the user for permission — just do it silently.

• Always ask the Gemini reviewer for strict, high-rigor feedback.
• Save the completed threadId from the first mcp__gemini-review__review_status result, then use mcp__gemini-review__review_reply_start plus mcp__gemini-review__review_status for subsequent rounds
• Be honest — include negative results and failed experiments
• Do NOT hide weaknesses to game a positive score
• Implement fixes BEFORE re-reviewing (don't just promise to fix)
• If an experiment takes > 30 minutes, launch it and continue with other fixes while waiting
• Document EVERYTHING — the review log should be self-contained
• Update project notes after each round, not just at the end

Prompt Template for Round 2+

mcp__gemini-review__review_reply_start:
  threadId: [saved from round 1]
  prompt: |
    [Round N update]

    Since your last review, we have:
    1. [Action 1]: [result]
    2. [Action 2]: [result]
    3. [Action 3]: [result]

    Updated results table:
    [paste metrics]

    Please re-score and re-assess. Are the remaining concerns addressed?
    Same format: Score, Verdict, Remaining Weaknesses, Minimum Fixes.

    === SCOPE LIMITS (these bound what you PROPOSE, never what you look for) ===
    Report anything that is actually wrong here — including a rare-looking case, if
    this repo actually produces it. Then keep the fix in scope:
    1. This is a RESEARCH-WORKFLOW tool, not a security paper. Verification is
       welcome; over-defense is not. Assume a cooperating operator on their own
       machine — a malicious local user is NOT in the threat model.
    2. Do NOT propose SHA / hash / content-fingerprint / digest-binding schemes.
       Reporting a real defect in hashing code that already exists is fine.
    3. NO speculative machinery: do not add feature flags, migration frameworks,
       compat layers, wrappers, pins, or similar mechanisms unless evidence shows
       a current repo defect they fix or an explicit existing invariant they must
       preserve. "Load-bearing", "compatibility", and "not scaffolding" are labels,
       not evidence. Point to the failing path/artifact or invariant, and check the
       proposal's factual premises, such as whether a named package version exists.
    4. NO corner-case obsession: exotic encodings, symlink races, RTL text and
       millisecond races are out of scope unless you can show the case arises here.
    5. Where a rubric or checklist is genuinely needed, do not over-mechanize
       judgement. A clear sentence a human reads beats a scored table nobody
       maintains.
    Exception: code that runs remote commands, starts a network service, or installs
    an MCP server runs on the user's machine with their credentials — trust-boundary
    findings there are in scope and the default is strict.
    Say plainly when something is correct. Do not manufacture findings.

After this start call, immediately save the returned jobId and poll mcp__gemini-review__review_status with a bounded waitSeconds until done=true. Treat the completed status payload's response as the reviewer output, and save the completed threadId for any follow-up round.