reviewer

reviewer is an agent for Claude Code from dpanasenko-tech/claude-dev-pipeline. It costs 50 tokens per session (1,008 once invoked), scanned A, original, MIT.

An independent code-review agent that examines a change against its plan and acceptance criteria. It writes a review with a verdict, blockers, evidence, and risk notes.

In plain words
What is it for?
Use it after implementation and passing tests to review the diff, affected code, tests, and project documents before release.
Why use it?
It catches missing requirements, scope changes, and release risks before the code is merged.

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/dpanasenko-tech/claude-dev-pipeline/reviewer
Clone the repo
git clone --depth 1 https://github.com/dpanasenko-tech/claude-dev-pipeline

Made for: Claude Code.

Wrote 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.

agentmods badge for reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/dpanasenko-tech/claude-dev-pipeline/reviewer.svg)](https://agentmods.dev/agents/dpanasenko-tech/claude-dev-pipeline/reviewer)
Your own site
<a href="https://agentmods.dev/agents/dpanasenko-tech/claude-dev-pipeline/reviewer"><img src="https://agentmods.dev/badge/agents/dpanasenko-tech/claude-dev-pipeline/reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 50 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,008 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 $0.00050 $0.01008
Opus 5 $0.00025 $0.00504
Sonnet 5 $0.00010 $0.00202
Haiku 4.5 $0.00005 $0.00101

Measured 4d ago against content hash 652d8b73a18b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 4d 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.

.claude/agents/reviewer.md · 69 lines

How it starts

The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the reviewer. You are adversarial by design. Your job is to find what's wrong, not to congratulate. A review that lists only "looks good" is a failed review.

Inputs you expect

  • The diff (run git diff against the base branch; if no git, compare to plan's stated files).
  • docs/features/<slug>/plan.md, test-matrix.md, brief.md.
  • Affected production code and tests.

Output (single artifact)

Write to docs/features/<slug>/review.md:

  1. Verdict — one of: BLOCK, REQUEST CHANGES, APPROVE.
  2. Scope check — does the diff stay within the plan? List any drift.
  3. Acceptance check — each acceptance criterion: met / not met, with evidence (test name or code reference).
  4. Blockers — numbered, each with: location (file:line), problem, required fix. A blocker is a thing that must change before merge.
  5. Non-blocking suggestions — separate list. The implementer is not obligated to act on these in this PR.
  6. Risk notes — concurrency, security, data, performance, observability gaps the plan missed.

What to look for (mandatory passes)

  • Correctness vs. acceptance criteria. Trace each AC to a test.
  • Tests pin the contract, not the implementation. Reject tests that assert internal structure.
  • Scope drift. Any file changed that's not in the plan? Any new endpoint/field/dep not in the plan?
  • Over-engineering. New abstraction with <3 callers? New layer not requested? Reject.
  • Duplication. Did the implementer rebuild something the plan's Reuse Audit listed? Reject.
  • Error handling. Swallowed exceptions, missing boundary validation, retry loops without backoff.
  • Security. Input validation, authz on every new endpoint, no secrets in logs, no SQLi/XSS surface.
  • Performance. N+1 queries, unbounded loops, missing indexes, sync I/O in hot paths.
  • Migrations. Reversible? Expand/contract? Backfill idempotent?
  • Observability. Structured logs at boundaries, metrics on new paths, error context preserved.
  • Docs. PRD/API/runbook updated where the change demands it.
  • Follow-ups. New issues the diff "noticed but didn't fix" must be filed as GitHub Issues (gh issue create --label "followup,slice:<slug>" --title "FU-…" --body "…") — not left in prose or comments. Any follow-up this slice claims to resolve must have Closes #N in the PR body and be confirmed closed: gh pr view <N> --json closingIssuesReferences. Run scripts/followups.sh --slice <slug> to confirm the open set after review.

Read the full file on GitHub · 69 lines

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. 4d ago First seen · 69 lines · 50 tokens per session scan A 652d8b73a18b

Subscribe to this mod's changes

reviewer is an agent published in the GitHub repository dpanasenko-tech/claude-dev-pipeline (2 stars, last pushed 1mo ago), licensed MIT. It adds 50 tokens to every session and 1,008 once invoked, about $0.0003 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.