rrr

rrr is a skill for Claude Code, Codex from dgk-dev/dgk-gpt. It costs 49 tokens per session (608 once invoked), scanned A, original, MIT.

A deeper code-review workflow that uses the GLM-5 model to examine a change and then verifies its findings against the actual code.

In plain words
What is it for?
Use it to review commits, staged changes, pull requests, or selected files.
Why use it?
It provides an additional, more thorough review while keeping only issues confirmed by inspection.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Codex.

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 skills/dgk-dev/dgk-gpt/rrr
Any agent
npx skills add dgk-dev/dgk-gpt --skill rrr
Clone the repo
git clone --depth 1 https://github.com/dgk-dev/dgk-gpt

Made for: Claude Code, Codex.

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 rrr

README.md
[![agentmods](https://agentmods.dev/badge/skills/dgk-dev/dgk-gpt/rrr.svg)](https://agentmods.dev/skills/dgk-dev/dgk-gpt/rrr)
Your own site
<a href="https://agentmods.dev/skills/dgk-dev/dgk-gpt/rrr"><img src="https://agentmods.dev/badge/skills/dgk-dev/dgk-gpt/rrr.svg" alt="Measured on agentmods" height="20"></a>
Per session 49 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 608 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 5.1 $0.00049 $0.00608
Opus 5 $0.00024 $0.00304
Sonnet 5 $0.00010 $0.00122
Haiku 4.5 $0.00005 $0.00061

Measured 6d ago against content hash a33e7bf7e0db, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

rrr 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 6d 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.

skills/rrr/SKILL.md · 85 lines

How it starts

The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/rrr

Use glm-review --model glm-5 as the deeper paid review pass, then validate the results against the code before you trust them.

Default Flow

  1. Determine the exact review scope first.
  • commit hash known -> commit mode
  • staged -> staged mode
  • pr -> PR mode
  • mixed unrelated local changes -> focused diff file
  1. Run a health check only when auth or connectivity looks questionable:
glm-review --model glm-5 --health
  1. Run the review with the smallest correct input:
glm-review --model glm-5
glm-review --model glm-5 --mode staged
glm-review --model glm-5 --mode pr
glm-review --model glm-5 --mode commit --ref <COMMIT_HASH>
glm-review --model glm-5 --diff-file /tmp/glm-review-diff.patch
  1. Verify every reported issue in the actual code and diff.
  2. Keep only confirmed issues.
  3. If fixes are requested or clearly in scope, apply them and rerun the closest verification.

Choosing Review Input

  • committed single change:
glm-review --model glm-5 --mode commit --ref <COMMIT_HASH>
  • committed subset of files:
glm-review --model glm-5 --mode commit --ref <COMMIT_HASH> --files src/a.ts src/b.ts
  • custom focused diff:
GIT_ROOT=$(git rev-parse --show-toplevel)
cd "$GIT_ROOT" && git diff HEAD -- <file1> <file2> ... > /tmp/glm-review-diff.patch
glm-review --model glm-5 --diff-file /tmp/glm-review-diff.patch

If the diff is empty, stop and say there is nothing to review.

Validation Rules

  • glm-review is a reviewer, not an oracle.
  • Re-check referenced code paths before reporting a bug.
  • Separate confirmed defects from debatable style feedback.
  • If a finding depends on broader context, inspect the relevant surrounding code before accepting it.

Error Handling

  • command not found: glm-review Install it with npm install -g glm-review
  • ZAI_API_KEY not set Export ZAI_API_KEY in the shell or your normal secret-loading path
  • auth or API failure Re-run glm-review --model glm-5 --health after refreshing credentials

Read the full file on GitHub · 85 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. 6d ago First seen · 85 lines · 49 tokens per session scan A a33e7bf7e0db

Subscribe to this mod's changes

rrr is a skill published in the GitHub repository dgk-dev/dgk-gpt (53 stars, last pushed 5mo ago), licensed MIT. It adds 49 tokens to every session and 608 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 2026-08-30.