agentmods.dev

chorus skill-files.instructions.md

A set of review rules for skill files used by Gemini and Codex, including required metadata, sections, commands, and output instructions. Skill files are instruction documents that tell an AI agent when and how to perform a task.

In plain words
What is it for?
Use it to review or prepare Gemini and Codex skill files, checking frontmatter, trigger descriptions, invocation commands, output handling, and required limitation notes.
Why use it?
It helps catch missing or incorrectly formatted parts before a skill is used. It also documents special requirements for file-reading skills and second-opinion tools.

Instructions file for GitHub Copilot

10 3mo ago A scan 395 tokens original MIT
View the source file valpere/chorus
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 instructions/valpere/chorus/skill-files
Clone the repo
git clone --depth 1 https://github.com/valpere/chorus

Made for: GitHub Copilot.

Per session 395 This file is loaded in full into every session.
When invoked 395 The same file — it is already loaded in full.
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 $0.00395 $0.00395
Opus 5 $0.00198 $0.00198
Sonnet 5 $0.00079 $0.00079
Haiku 4.5 $0.00040 $0.00040

Measured 3d ago against content hash 7c1c2db3b80c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

chorus skill-files.instructions.md 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.

.github/instructions/skill-files.instructions.md · 32 lines

What it actually says

SKILL.md review guidelines

Gemini skills (for-gemini/) must have frontmatter with:

  • name — kebab-case, prefixed with chorus-
  • description — multi-line prose with trigger phrases the model will match on
  • license: MIT
  • metadata.author and metadata.version

Codex skills (for-codex/) must have frontmatter with only:

  • name — kebab-case, prefixed with chorus-
  • description — single concise line

All skills must include:

  • ## When to use — bullet list of trigger phrases and scenarios
  • ## Invocation — bash code block with the exact CLI command(s) to run
  • ## Output handling — what to do with the result (verbatim / synthesize / etc.)

Codex skills that involve file reading (parallel-review, parallel-debug) must include a ## Known limitation section noting the sandbox restriction.

Second-opinion defaults are host-specific: a host's own skill should not default to itself. Example: the chorus command defaults to Gemini; for-gemini/second-opinion defaults to Claude (Gemini is the caller, so it asks someone else).

Invocation bash blocks must use the correct non-interactive flags:

  • Claude: claude --print "..." --dangerously-skip-permissions
  • Gemini: gemini --prompt "..." --yolo --output-format text
  • Codex: codex exec "..."
  • Cursor: agent -p --force "..." (binary is agent, not cursor)
  • Kilo: kilo run --auto "..."
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 · 32 lines · 395 tokens per session scan A 7c1c2db3b80c

Subscribe to this mod's changes

chorus skill-files.instructions.md is an instructions file published in the GitHub repository valpere/chorus (10 stars, last pushed 3mo ago), licensed MIT. It adds 395 tokens to every session, about $0.0020 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.

Building agentsai-toolsclaude-codecodexdeveloper-toolsgemini-cli
Related

Other instructions, from other repositories

TokenPilot AGENTS.md

AGENTS.md instructions for eowlsdh/TokenPilot, covering project knowledge base, structure, where to look, code map and conventions.

eowlsdh/TokenPilot · 1,618 tokens

Project-AMBER AGENTS.md

Instructions for JJHbrams/Project-AMBER, covering agent orchestration, roles and non-negotiable rules.

JJHbrams/Project-AMBER · 400 tokens

TokenPilot CLAUDE.md

Claude Code instructions for eowlsdh/TokenPilot, a project described as: Local-first macOS menu bar utility for AI coding-tool usage metadata.

eowlsdh/TokenPilot · 91 tokens

Skills AGENTS.md

AGENTS.md instructions for AceDataCloud/Skills, covering acedatacloud agent skills, available skills, ai music & audio, ai image generation and ai video generation.

AceDataCloud/Skills · 647 tokens

Skills copilot-instructions.md

Copilot instructions for AceDataCloud/Skills, covering copilot instructions for skills, sync from docs, project structure and important.

AceDataCloud/Skills · 450 tokens

semble-hooks CLAUDE.md

Instructions for benediktkraus/semble-hooks, covering semble hooks — claude.md, was ist das, projekt-typ, tech-stack and architektur-referenz.

benediktkraus/semble-hooks · 984 tokens