magent-refine

magent-refine is a command for Claude Code from gobing-ai/superskill. It costs 10 tokens per session (611 once invoked), scanned A, original, Apache-2.0.

A tool for reviewing and improving an agent configuration file such as AGENTS.md or CLAUDE.md, which contains instructions for coding tools.

In plain words
What is it for?
Use it to evaluate a configuration, preview or apply supported fixes, recheck the result, and save the evaluation for later analysis.
Why use it?
It identifies configuration issues and suggests improvements while accounting for the fact that these files may be plain Markdown without required metadata.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions CLAUDE.md; mentions Claude Code.

Part of the cc plugin — 6 skills, 17 commands, 5 agents shipped together

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 commands/gobing-ai/superskill/magent-refine
Clone the repo
git clone --depth 1 https://github.com/gobing-ai/superskill

Made for: Claude Code.

Or install cc, the plugin that ships this one along with the rest of its 6 skills, 17 commands, 5 agents.

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 magent-refine

README.md
[![agentmods](https://agentmods.dev/badge/commands/gobing-ai/superskill/magent-refine.svg)](https://agentmods.dev/commands/gobing-ai/superskill/magent-refine)
Your own site
<a href="https://agentmods.dev/commands/gobing-ai/superskill/magent-refine"><img src="https://agentmods.dev/badge/commands/gobing-ai/superskill/magent-refine.svg" alt="Measured on agentmods" height="20"></a>
Per session 10 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 611 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 $0.00010 $0.00611
Opus 5 $0.00005 $0.00305
Sonnet 5 $0.00002 $0.00122
Haiku 4.5 $0.00001 $0.00061

Measured 2d ago against content hash 8648c1284eb4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

magent-refine 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 2d 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.

plugins/cc/commands/magent-refine.md · 74 lines

How it starts

The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Magent Refine

Wraps cc:cc-magents skill.

Run evaluation, apply deterministic structural fixes (missing fields, type coercion, whitespace), then re-evaluate. Magents are frontmatter-optional plain markdown, so --auto is usually a clean no-op on their body: it surfaces low-scoring dimensions but does not perform semantic rewrites. The invoking agent owns those edits under the cc:cc-magents workflow.

When to Use

  • Improve magent config quality after scaffolding
  • Apply low-risk structural fixes and surface semantic work
  • Preview fixes without writing with --dry-run

Arguments

Argument Description Default
<nameOrPath> Configuration name or path to its .md file (required)
--auto Skip interactive prompts (auto-apply fixes) false
--save Persist the evaluation to the evaluation store false
--dry-run Preview classified fixes and projected delta without writing false
--target Target platform claude

Examples

# Refine a config (evaluate + structural fixes + re-evaluate)
/cc:magent-refine ./CLAUDE.md
# Auto-refine without prompts
/cc:magent-refine ./CLAUDE.md --auto --save
# Preview fixes without writing
/cc:magent-refine ./CLAUDE.md --dry-run

Refinement Contract

Follow cc:cc-magents workflows § "Main-Agent Evaluation and Refinement". Freeze requested layout and critical boundaries; fix contradictions and dead commands first; delete no-ops and sediment; collapse duplicate owners; move rarely needed depth to live authoritative links; add only missing non-inferable guidance. Preserve safety and verification inline on targets that do not receive rule modules.

After semantic edits, assemble every target and re-run strict validation plus evaluation. Compare every dimension with baseline, not only the aggregate. Do not claim --auto applied a body change that it only suggested.

Implementation

Pass $ARGUMENTS to the underlying skill for processing.

Read the full file on GitHub · 74 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. 2d ago Changed · +7 lines 8648c1284eb4
  2. 6d ago First seen · 67 lines · 10 tokens per session scan A 4246f705c202

Subscribe to this mod's changes

magent-refine is a command published in the GitHub repository gobing-ai/superskill (5 stars, last pushed 3d ago), licensed Apache-2.0. It adds 10 tokens to every session and 611 once invoked, about $0.0001 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.