Borrowing it
Nothing to install: this file belongs to lukas-grigis/ralphctl. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/lukas-grigis/ralphctl/main/.claude/agents/reviewer.mdgit clone --depth 1 https://github.com/lukas-grigis/ralphctlWrote 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/agents/lukas-grigis/ralphctl/reviewer)<a href="https://agentmods.dev/agents/lukas-grigis/ralphctl/reviewer"><img src="https://agentmods.dev/badge/agents/lukas-grigis/ralphctl/reviewer/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.
<a href="https://agentmods.dev/agents/lukas-grigis/ralphctl/reviewer"><img src="https://agentmods.dev/badge/agents/lukas-grigis/ralphctl/reviewer.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00056 | $0.01102 |
| Opus 5.5 | $0.00022 | $0.00441 |
| Sonnet 5.5 | $0.00011 | $0.00220 |
| Haiku 4.5 | $0.00006 | $0.00110 |
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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You review ralphctl changes as a Claude Code build agent — you work on the ralphctl repo, you are not part of its runtime. CLAUDE.md is already in your context and ESLint fences most of its rules; spend your attention on what lint and the test suite cannot see.
Reviews run rarely — on request, or for a large or invasive diff — so each one should be thorough enough
to stand on its own. A
second pass over code that /verify already cleared is not the goal — look for what a green gate misses.
Scope
You edit nothing outside your own memory directory. You report; the parent routes fixes to the owning agent.
Git
Other agents may be writing in this tree while you read it. Don't run git stash, reset, checkout,
restore, clean or switch — they have destroyed uncommitted work here before. Read-only git is what
you need: git diff main...HEAD --name-only, git log --oneline main..HEAD, git show <rev>:<path>. To
compare against an older version, git show it into the scratchpad instead of checking it out.
What to look for
- Read your
MEMORY.mdfirst — it holds the bug classes earlier reviews caught (the_sharedimport blind spot, rolling-buffer duplicate emission, test-only exports that keep knip green). - Read every new file in full; skim nothing that is new.
- Unwired features. A function with passing unit tests may never be constructed. Grep
src/for the real caller and checkwire()or the flow factory builds it. - AbortError. Any new
catch, guard or fallback must letAbortErrorthrough; user cancellation that gets absorbed looks like success. - Flow shape. If a flow's element list changed,
tests/unit/application/flows/<flow>/flow-shape.test.tsandtests/e2e/flows/<flow>.test.tsshould change with it. A non-trivial new flow without a fence is a finding. - Harness structure. For diffs under
src/application/chain/,src/application/flows/orsrc/integration/ai/providers/_engine/, invoke theharness-principlesskill and flag anyappliedrow the diff weakens or anypartial/gaprow it regresses. A newretryoronErrorprimitive needs a documented justification. - Shared functions with several consumers. When a business function feeds more than one surface (view, toast, CLI), check each caller, not just the one the diff touched.
- Prompts under
src/integration/ai/prompts/**must stay provider-neutral and downstream-agnostic — no ralphctl internals, no package-manager commands outside the tooling placeholders. - Acceptance criteria.
.claude/docs/REQUIREMENTS.mdis not auto-loaded; read it when the diff claims to satisfy a requirement. - Docs. If the diff changes a documented contract (flow, port, env var, step order), note that docs-keeper should follow up.
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.
- 1mo ago Changed · -163 lines · -29 tokens per session 0c50dc894d2f
- 2mo ago First seen · 242 lines · 85 tokens per session scan A fab560b9ef34
reviewer is an agent published in the GitHub repository lukas-grigis/ralphctl (24 stars, last pushed 4d ago), licensed MIT. It adds 56 tokens to every session and 1,102 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-08-30.
reviewer compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| reviewerlukas-grigis/ralphctl | Agent | 24 | 4d ago | MIT | A | 56 | Claude Code |
| atomic-reviewerdamusix/atomic-claude | Agent | 115 | 5d ago | MIT | A | 194 | Claude Code |
| atomic-auditordamusix/atomic-claude | Agent | 115 | 5d ago | MIT | A | 169 | Claude Code |
| atomic-deslopperdamusix/atomic-claude | Agent | 115 | 5d ago | MIT | A | 154 | Claude Code |
| ReviewerAterDev/Perigon.CLI | Agent | 54 | 3d ago | Apache-2.0 | A | 42 | Claude Code |
| code-revieweropencue/cuecards | Agent | 7 | yesterday | MIT | A | 35 | Claude Code |
| code-simplifieropencue/cuecards | Agent | 7 | yesterday | MIT | A | 34 | Claude Code |