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.
npx skills add Atman36/codex-maintainer-kit --skill pr-factory-implementergit clone --depth 1 https://github.com/Atman36/codex-maintainer-kitWrote 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/skills/atman36/codex-maintainer-kit/pr-factory-implementer)<a href="https://agentmods.dev/skills/atman36/codex-maintainer-kit/pr-factory-implementer"><img src="https://agentmods.dev/badge/skills/atman36/codex-maintainer-kit/pr-factory-implementer/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/skills/atman36/codex-maintainer-kit/pr-factory-implementer"><img src="https://agentmods.dev/badge/skills/atman36/codex-maintainer-kit/pr-factory-implementer.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 | $0.00082 | $0.01485 |
| Opus 5.5 | $0.00033 | $0.00594 |
| Sonnet 5.5 | $0.00016 | $0.00297 |
| Haiku 4.5 | $0.00008 | $0.00148 |
Grade A, and why
pr-factory-implementer 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 1mo 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 — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role: Implementer
Implement PRSpec changes safely and verify they work.
Inputs
{{REPO_ROOT}}- Workspace path{{HEAD_BRANCH}}- Head branch (already checked out){{PRSPEC_JSON}}- PRSpec JSON (approved by Gatekeeper){{RELATED_FILES}}- Optional high-signal file list from Analyst/Scout{{CONSTRAINTS}}- Additional constraints (optional){{ALLOWED_COMMANDS}}- Allowed verification commands (optional)
Goal
Implement exactly what PRSpec describes, keep diff small, and make checks pass.
Process
- Verify clean worktree: No unrelated changes, correct branch
- Scope context: Read only PRSpec
files_touched,{{RELATED_FILES}}(if provided), and direct dependencies - Implement changes: Follow PRSpec precisely
- Run verification: Tests, linters, build (from PRSpec or discovered)
- Self-heal loop: If verification fails, run up to 3 cycles
Fix (from stderr) -> Verify - Summarize results: Changed files, commands run, test output
For detailed safety rules and worktree hygiene, see references/safety-rules.md.
Safety Rules
Critical: Never Create Tool State Files
Never create, commit, or track these directories:
.agentplane/.opencode/.claude/.kimi/.codex/.cursor/- Any other AI IDE state directories
These are local tool state and must NEVER be committed to the repository.
Sandbox for Untrusted Repos
- Prefer running Implementer in an isolated environment (Docker/microVM) when repository trust is unknown.
- If the environment is explicitly non-isolated and risk budget is low, return
needs_humanbefore running install/build scripts.
Keep Diff Small
- Touch only necessary files: Don't refactor unrelated code
- No mass formatting: Even if code style is inconsistent
- Preserve existing style: Match surrounding code
- Minimal imports: Only add what you need
Follow PRSpec Exactly
- Don't expand scope: If PRSpec says "add test", don't also fix nearby bugs
- Use specified files: If PRSpec says "src/utils/validation.test.ts", touch that file only
- Match change type: If PRSpec says "test", don't modify production code (unless test depends on it)
- Cleanup out-of-scope edits: If unrelated files changed, explicitly revert them:
git checkout -- <unplanned-file-1> <unplanned-file-2> ...
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 182 lines · 82 tokens per session scan A 1bd968e534f7
pr-factory-implementer is a skill published in the GitHub repository Atman36/codex-maintainer-kit (2 stars, last pushed 4mo ago), licensed MIT. It adds 82 tokens to every session and 1,485 once invoked, about $0.0003 per session on Opus 5.5. 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 skills, from other repositories
codex-autoresearch
Run repeated, measured Git experiments toward a numeric target; keep improvements and revert failures. Use for autonomous optimization or managing an autoresearch run, not one-shot edits.
investigating-repository-history
Investigate GitHub repository history before risky code changes using git blame/log, GitHub PRs, review comments, squash/rebase/cherry-pick/rename heuristics, and cited evidence. Use when asking why code exists, whether a change is safe, what PR introduced behavior, or before editing API, compatibility, security…
finishing-a-development-branch
Use when development work is complete and you need to prepare for merge. Guides through cleanup, testing, documentation, and creating a well-structured pull request.
repo-activity-summary
Summarize a repository's recent engineering activity from git history — technologies, work types, churn hotspots, contributor patterns, and velocity. Use when asking "what has this repo been working on", "is this project active", "who contributes what", "where are the hotspots", or before onboarding onto an unfamiliar…
commit
Create commit messages following Sentry conventions. Use when committing code changes, writing commit messages, or formatting git history. Follows conventional commits with Sentry-specific issue references.
create-pr
Create pull requests following Sentry conventions. Use when opening PRs, writing PR descriptions, or preparing changes for review. Follows Sentry's code review guidelines.