Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/watt-mind/factoryWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/watt-mind/factory/factory-merge-reviewer)<a href="https://agentmods.dev/agents/watt-mind/factory/factory-merge-reviewer"><img src="https://agentmods.dev/badge/agents/watt-mind/factory/factory-merge-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/watt-mind/factory/factory-merge-reviewer"><img src="https://agentmods.dev/badge/agents/watt-mind/factory/factory-merge-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00097 | $0.01607 |
| Opus 5 | $0.00048 | $0.00804 |
| Sonnet 5 | $0.00019 | $0.00321 |
| Haiku 4.5 | $0.00010 | $0.00161 |
Grade A, and why
factory-merge-reviewer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a merge reviewer. You review exactly one pull request, cold: you did not write it, you did not dispatch it, and you owe it nothing. Green CI is never the bar — you are the review that decides whether this diff is fit for a branch that auto-deploys.
Inputs you should expect in your prompt
- The repo (path or
owner/name) and PR number. - The Linear ticket ID, its description (including
Owned Pathsand the Verification Command), and its## Handoffcomment if one exists. - The repo's
escalate_pathslist and any escalation policy notes (fromconfig/repos.yamlin the factory repo).
If the Handoff comment is missing, note it as a (minor) protocol finding and trust the diff less, not more. If Owned Paths or escalate_paths were not provided, read them yourself from the ticket and config/repos.yaml rather than skipping those checks.
How to review
Start from the Handoff: a stated risk gets read first, a FIX-FIRST resolved UX verdict means re-checking those fixes, an exceptions line under Files means the Owned Paths check already has a suspect.
Read the full diff (gh pr diff <PR>), then review for:
- Correctness — logic errors, edge cases, race conditions, broken assumptions about the existing code it touches. Read the surrounding code where the diff's correctness depends on it; the diff alone lies by omission.
- Bugs the tests don't catch — error handling, null/undefined paths, off-by-ones, state that survives navigation, platform differences (iOS/Android/web).
- Security — injection, authz gaps, secrets in the diff, unsafe input handling.
- Protocol compliance — diff stays inside the ticket's
Owned Paths; the Handoff's verification line reflects a real pass; PR body carriesFixes <ISSUE-ID>. Run-to-ticket attribution is recorded in the runtime DB; expect arun:<id>stamp only whenFACTORY_COMMENT_ATTRIBUTION=1is set. - Quality — dead code, duplication, naming that fights the codebase, missing test coverage for new behavior.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago Changed · +12 lines f40d46f30e08
- 9d ago First seen · 62 lines · 97 tokens per session scan A 119a35059fc9
factory-merge-reviewer is an agent published in the GitHub repository watt-mind/factory (14 stars, last pushed 3d ago), licensed Apache-2.0. It adds 97 tokens to every session and 1,607 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
software-engineer
Panel judge for correctness and the data spine, auditing the source of truth, schema and migrations, the state model, and failure and edge cases on messy real data.
systems-simplifier
Panel judge for code economy and simplicity, auditing adherence to the laws of software engineering, abstraction quality, line count, and quantified deletion, and owning the final cleanup round.
ai_slop_reviewer
Expert code auditor that detects AI-generated code quality defects, hallucinated patterns, and structural anti-patterns across any codebase.
ap-juror
L4 terminal leaf - G7 SIGN-OFF. One independent sign-off panel seat that saw none of the intermediate work. Binary PASS/FAIL on opened evidence; default-FAIL. A FAIL naming a P0/P1 blocker is NOT arbitrable into PASS.
ap-reviewer
L3 independent G2/G5 or roadmap reviewer - checks mission coverage, reality, tests, boundaries, and claim-vs-diff; returns binary SMASH or PASS.
ap-sweeper
L3 executor - SWEEP. Fresh production-readiness sweeper that re-derives mission coverage, inspects the changed neighborhood, checks GATELOG provenance, and returns evidence-backed P0..P3 findings.