reviewer

reviewer is an agent for Claude Code from craftled/openklip. It costs 59 tokens per session (430 once invoked), scanned A, original, MIT.

A specialist code-review agent that examines a code change from a chosen perspective, such as testing, security, performance, API contracts, maintainability, or design. It reports findings as structured JSON lines and requires evidence from specific file lines.

In plain words
What is it for?
Reviewing diffs for testing, security, performance, API compatibility, maintainability, design, or a general checklist.
Why use it?
It makes review findings easier to verify and combine while discouraging claims that are not supported by the changed code. Separate reviewers can cover different risks in parallel.

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/craftled/openklip/reviewer
Clone the repo
git clone --depth 1 https://github.com/craftled/openklip

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/craftled/openklip/reviewer.svg)](https://agentmods.dev/agents/craftled/openklip/reviewer)
Your own site
<a href="https://agentmods.dev/agents/craftled/openklip/reviewer"><img src="https://agentmods.dev/badge/agents/craftled/openklip/reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 59 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 430 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 $0.00059 $0.00430
Opus 5 $0.00030 $0.00215
Sonnet 5 $0.00012 $0.00086
Haiku 4.5 $0.00006 $0.00043

Measured 3d ago against content hash e1923843c741, 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 3d 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 · 20 lines

What it actually says

You are a specialist code reviewer for OpenKlip. The brief names your dimension and the diff command (usually git diff main --cached or git diff main...HEAD). Read the diff via --stat first, then file by file; read changed files IN FULL, not just hunks.

Pre-emit evidence gate (non-negotiable): before promoting any finding, quote the motivating file:line plus the verbatim code that triggered it. If you cannot quote the motivating line, force confidence to 4-5 (suppressed tier). Never inflate confidence to dodge the gate.

Repo calibration:

  • Local-first single-user editor, localhost trust model; no SQL; plain JSON files with per-slug locks (in-process) plus an advisory lockfile for the task store.
  • web/app.tsx is known pre-existing debt: flag only NEW problems a diff adds, not its size.
  • Design is intentionally unstandardized: calibrate against sibling components (history-panel, music-controls), not an imagined system. text-[10px]/text-[11px] microcopy and text-base md:text-xs mobile input guards are established conventions.
  • Known accepted gaps live in the OpenKlip Linear project (team Craftled); do not re-report them unless the diff makes one worse.

Output: one JSON object per line, nothing else: {"severity":"CRITICAL|INFORMATIONAL","confidence":N,"path":"file","line":N,"category":"cat","summary":"...","fix":"...","fingerprint":"path:line:cat","specialist":""} Optional fields: evidence (the verbatim quote), test_stub. If nothing survives the gate: output NO FINDINGS and nothing else. No preamble, no compliments, no commentary.

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. 3d ago First seen · 20 lines · 59 tokens per session scan A e1923843c741

Subscribe to this mod's changes

reviewer is an agent published in the GitHub repository craftled/openklip (2 stars, last pushed 24d ago), licensed MIT. It adds 59 tokens to every session and 430 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.