claude-code-power-config: Agent for Claude Code

.claude/agents/review-validator.md

review-validator is an agent for Claude Code from AlexisBalayre/claude-code-power-config. It costs 42 tokens per session (633 once invoked), scanned A, original, MIT.

A separate checking agent that tests whether a proposed code-review finding is actually supported by the code. It tries to disprove the finding before it is posted or acted on.

In plain words
What is it for?
Use it in an automated pull-request review process to validate reported bugs, security concerns, and convention violations against the repository.
Why use it?
It reduces false bug reports caused by assumptions, inaccurate references, or rules that do not apply to the changed code. This gives review findings an additional verification step.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; reads .claude/ paths; mentions CLAUDE.md.

This is AlexisBalayre/claude-code-power-config's own configuration. It tells Claude Code how to work on claude-code-power-config itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything claude-code-power-config configures →

Reuse

Borrowing it

Nothing to install: this file belongs to AlexisBalayre/claude-code-power-config. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/AlexisBalayre/claude-code-power-config/main/.claude/agents/review-validator.md
Clone the repo
git clone --depth 1 https://github.com/AlexisBalayre/claude-code-power-config

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 review-validator

README.md
[![agentmods](https://agentmods.dev/badge/agents/alexisbalayre/claude-code-power-config/review-validator/github.svg)](https://agentmods.dev/agents/alexisbalayre/claude-code-power-config/review-validator)
Your own site
<a href="https://agentmods.dev/agents/alexisbalayre/claude-code-power-config/review-validator"><img src="https://agentmods.dev/badge/agents/alexisbalayre/claude-code-power-config/review-validator/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.

agentmods 80×15 button for review-validator

Your own site · 80×15
<a href="https://agentmods.dev/agents/alexisbalayre/claude-code-power-config/review-validator"><img src="https://agentmods.dev/badge/agents/alexisbalayre/claude-code-power-config/review-validator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 42 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 633 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00042 $0.00633
Opus 5 $0.00021 $0.00316
Sonnet 5 $0.00008 $0.00127
Haiku 4.5 $0.00004 $0.00063

Measured 8d ago against content hash c2ab23c8b2f2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

review-validator 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 8d 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/review-validator.md · 30 lines

How it starts

The opening of the file, as written. The whole thing — 30 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Validation gate

You are the inverse of a reviewer. The orchestrator hands you one finding (or a batch of minor ones) that is about to be acted on; your job is to try to refute it against the actual code. A finding posted or fixed on a false premise erodes trust in the whole review, so be genuinely adversarial: read the code the finding points at, follow the references its claim relies on, and ask whether the issue is really there as described.

A finding is confirmed only when the code, read directly, shows the issue as described. Imprecise details do not refute it: a slightly-off line number or file path is a correction, not a refutation. But a finding whose core claim does not hold (the bug cannot happen, the rule is not actually violated, the prose is accurate after all) is refuted no matter how plausible it sounds.

Rule authority: a style or convention rule can legitimately live in a reviewer manifest (.claude/agents/review-*.md) rather than in repo docs; the manifests are standing review policy. When a finding cites one, read that manifest and adjudicate scope (is the flagged text something the change adds, inside the rule's stated bounds?) instead of refuting the rule as fabricated because docs/conventions and CLAUDE.md do not state it. A rule found in neither the cited manifest nor the repo docs is still fabricated.

When reading leaves the verdict open, settle it with a spike: execute the code path with the input the finding worries about (a one-liner, a targeted test) and let the result decide. Keep spikes throwaway and trace-free: leave the tree and its state exactly as you found them.

Not under CI. When your brief says the run is CI, the review is static: the checkout has no node_modules, and the command guard blocks execution, redirection, and temp files, so every spike attempt burns tool calls to end where static reading starts. Settle open verdicts by reading the checked-out tree and the pinned dependency source directly; a claim about a dependency's internals you cannot read there rests on pinned-version knowledge and caps at medium confidence.

You judge only what you are given: do not re-scope, soften, re-tag, or hunt for new issues.

Read the full file on GitHub · 30 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. 8d ago First seen · 30 lines · 42 tokens per session scan A c2ab23c8b2f2

Subscribe to this mod's changes

review-validator is an agent published in the GitHub repository AlexisBalayre/claude-code-power-config (2 stars, last pushed 28d ago), licensed MIT. It adds 42 tokens to every session and 633 once invoked, about $0.0002 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.

Related

Other agents, from other repositories

code-reviewer

A code-review agent that checks whether changes follow their specification and assesses code quality, security, maintainability, and performance. It reports findings with severity levels and file-and-line references.

sangrokjung/claude-forge · 96 tokens

adversarial-reviewer

Independent read-only checker for behavioural changes. Runs in a fresh context that did not author the change, reproduces the claim against the goal, spec, diff and execution evidence, and returns exactly one verdict — APPROVE, REQUESTCHANGES or UNVERIFIED — as a forge.review/v1 envelope. MUST BE USED before claiming…

sangrokjung/claude-forge · 102 tokens

security-reviewer

A read-only security review agent that checks code for common web risks, exposed secrets, unsafe input handling, authentication and authorization problems, and dependency issues. OWASP Top 10 is a widely used list of major web application security risks.

sangrokjung/claude-forge · 95 tokens

database-reviewer

Use when writing SQL queries, creating migrations, or troubleshooting database performance in Supabase/PostgreSQL projects. Reviews indexes, RLS policies, schema types, N+1 patterns. Read-only reviewer with EXPLAIN ANALYZE capability.

sangrokjung/claude-forge · 52 tokens

refactor-cleaner

An agent for finding and safely removing dead code, unused exports, unused dependencies, and duplicate implementations.

sangrokjung/claude-forge · 109 tokens

skeptical-auditor

Independent skeptical re-verification after verify-agent (or any self-verifying agent) claims a pass. Read-only and adversarial: re-runs every step that was claimed, compares actual exit codes against the claim, and is paid to find failures rather than confirm success. Never approves without executed evidence. Spawned…

sangrokjung/claude-forge · 120 tokens