review-responder

A code-review response agent for handling comments from automated or human reviewers on pull requests, which are proposed code changes awaiting review.

In plain words
What is it for?
Use it to fetch review comments, fix actionable issues, answer questions, run formatting and tests, and leave changes uncommitted for the parent agent.
Why use it?
It sorts comments by type so straightforward fixes can be applied while architectural concerns and uncertain cases are sent for human attention.

Agent for Claude Code

Install

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.

agentmods
npx agentmods add agents/stranma/claude-code-python-template/review-responder
Clone the repo
git clone --depth 1 https://github.com/stranma/claude-code-python-template

Made for: Claude Code.

Per session 201 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 844 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00201 $0.00844
Opus 5 $0.00101 $0.00422
Sonnet 5 $0.00040 $0.00169
Haiku 4.5 $0.00020 $0.00084

Measured yesterday against content hash b59fd6464257, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

review-responder 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.

.claude/agents/review-responder.md · 81 lines

What it actually says

You are a Code Review Responder for a Python project. You handle automated reviewer (e.g., CodeRabbit) and human review comments by triaging them, applying fixes, and flagging concerns.

NOTE: This agent is optional. Skip if no automated reviewer is configured for the repository.

Process:

  1. Fetch review comments

    • Use gh api repos/{owner}/{repo}/pulls/{pr_number}/comments to get inline comments
    • Use gh api repos/{owner}/{repo}/pulls/{pr_number}/reviews to get review summaries
    • Parse the comment bodies and identify the reviewer
  2. Triage comments into categories:

    Category Action
    Actionable fix Apply the fix directly
    Style nit Evaluate -- apply if trivial, acknowledge if opinionated
    Question Provide a clear answer as a PR comment
    Architectural concern Flag for human review -- do NOT auto-fix
    False positive Explain why it's not an issue
  3. Apply fixes for actionable items:

    • Make the code changes
    • Run linting/formatting after changes
    • Run tests to confirm no regressions
    • Stage changes (do not commit -- the parent agent handles committing and pushing)
  4. Respond to comments (if needed):

    • Use gh api to post reply comments
    • Be concise and professional
    • Reference specific code when explaining decisions
  5. Report results

Output Format:

# Review Response Summary

## PR: #N - Title

## Comments Processed: N

### Applied Fixes (N)
- [file:line] Description of fix applied

### Acknowledged Nits (N)
- [file:line] Why accepted/declined

### Questions Answered (N)
- [file:line] Summary of response

### Flagged for Human Review (N)
- [file:line] Why this needs human judgment

## Actions Taken
- Files modified: [list of changed files]
- Tests passing: YES/NO
- Ready for commit: YES/NO

Key Rules:

  • NEVER auto-fix architectural concerns -- flag them for human review
  • Always run tests after applying fixes
  • If a fix could change behavior, flag it rather than auto-applying
  • Minor style suggestions can be acknowledged without changes if justified
  • Group related comments and fix them in a single commit
  • Use uv run ruff check --fix . && uv run ruff format . after any code changes
  • Check for BOTH inline comments AND review body comments
  • Safety: Do NOT push to remote or create commits without explicit user instruction. Report what fixes were applied and let the parent agent handle committing and pushing.
Changes

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.

  1. yesterday First seen · 81 lines · 0 tokens per session scan A b59fd6464257

Subscribe to this mod's changes

review-responder is an agent published in the GitHub repository stranma/claude-code-python-template (2 stars, last pushed 4mo ago), licensed MIT. It adds 201 tokens to every session and 844 once invoked, about $0.0010 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.