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 agentmods add commands/pablomarin/claude-codex-forge/review-pr-commentsgit clone --depth 1 https://github.com/pablomarin/claude-codex-forgeWrote 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/commands/pablomarin/claude-codex-forge/review-pr-comments)<a href="https://agentmods.dev/commands/pablomarin/claude-codex-forge/review-pr-comments"><img src="https://agentmods.dev/badge/commands/pablomarin/claude-codex-forge/review-pr-comments.svg" alt="Measured on agentmods" 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 | $0.00000 | $0.00518 |
| Opus 5 | $0.00000 | $0.00259 |
| Sonnet 5 | $0.00000 | $0.00104 |
| Haiku 4.5 | $0.00000 | $0.00052 |
Grade A, and why
review-pr-comments 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 yesterday.
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 — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/review-pr-comments — Host-Neutral Review Feedback
Use after a PR receives automated or human feedback. The active Claude Code or Codex host remains the main agent for this session; reviewer choice is independent.
1. Read the Current Review Set
Resolve the current PR and fetch its review summaries, inline comments, unresolved conversations,
base, and head SHA. If no PR exists, stop. Record the review-set fingerprint and current candidate
in .forge/local/state.md; never treat stale comments or a previous candidate receipt as current.
2. Evaluate Before Editing
For each comment, identify the requested change and verify it against code, requirements, and project rules. Do not agree performatively:
- valid finding — name the affected behavior and smallest correction;
- already fixed/stale — cite the current evidence;
- incorrect or conflicting — explain the technical objection;
- ambiguous or consequential — dispatch the Forge opinion workflow with the active host entry point
(
/opinionin Claude Code;$opinionin Codex), relevant artifact, and question.
Use automatic other-engine review with visible fresh same-engine fallback. A finding is not an engine failure and never triggers fallback.
Treat the current fingerprinted review set as one broad review, use one repair pass, then one closure review limited to named findings and direct regressions. Do not start a second broad scan for unchanged comments/candidate. One still-open reachable P0/P1 may receive one surgical repair and verification; then surface the blocker to the developer. P3, cosmetic, and speculative concerns do not keep the loop open.
3. Repair and Re-Certify
Apply accepted fixes with TDD where behavior changes, then run focused owning checks. Run the
Forge-owned simplification phase for code mutations. Stage all intended changes, freeze a new
candidate, and rerun the final gates that the mutation can affect: distinct code-spec and
code-quality reviews, verify-app, and applicable E2E. Persist candidate-bound receipts under
.forge/local/; any later mutation invalidates affected evidence.
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.
- yesterday Changed · -36 lines b23163c42378
- 5d ago First seen · 82 lines · 0 tokens per session scan A 8d40cf35857e
review-pr-comments is a command published in the GitHub repository pablomarin/claude-codex-forge (5 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 518 tokens. 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 commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.