Is the Agent Instructions skill free?

Yes. The skill itself is free from nimadorostkar/Claude-Skills-collection. SkillProof publishes the install command and an independent test verdict at no cost.

Does Agent Instructions work with Claude Code?

We tested it with Claude Code 2.x (agent harness) on Aug 7, 2026. Verdict: Tested · Works. Wrote a CLAUDE.md for the consensus-loop repo twice and then executed every command in both files. The no-skill version shipped five commands of which three were broken, including `node --test tests/` — the exact command the repo's real CLAUDE.md carries, which dies on Node 26 with 'Cannot find module /tests'. Following the skill's rule that commands must be run before they are written turned two of those traps into explicit do-not-run lines with the real error text, and surfaced a third: `npm run quality` in test-harness cannot pass because @types/node is missing, so `npx vitest run` (44/44 green) replaced it. Cost of the better file was 59 lines against 39.

What is the Agent Instructions SkillProof Score?

9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.

How do I install Agent Instructions?

Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.

Can I use Agent Instructions with Cursor, Copilot, Gemini CLI, Codex or other AI tools?

The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.

Agent Instructions

Writes a CLAUDE.md/AGENTS.md that lists only verified commands and real gotchas

by nimadorostkar · nimadorostkar/Claude-Skills-collection

Tested · Works

Install — copy the command Download source ↗

Test report

Verdict: Tested · Works
Score: 9.2/10
Tested: Aug 7, 2026
Environment: Claude Code 2.x (agent harness)

Wrote a CLAUDE.md for the consensus-loop repo twice and then executed every command in both files. The no-skill version shipped five commands of which three were broken, including `node --test tests/` — the exact command the repo's real CLAUDE.md carries, which dies on Node 26 with 'Cannot find module <repo>/tests'. Following the skill's rule that commands must be run before they are written turned two of those traps into explicit do-not-run lines with the real error text, and surfaced a third: `npm run quality` in test-harness cannot pass because @types/node is missing, so `npx vitest run` (44/44 green) replaced it. Cost of the better file was 59 lines against 39.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

Installs cleanly 5/5
Triggers reliably 5/5
Output vs. baseline 8/10
Docs & honesty 5/5

What Agent Instructions does

Guidance for writing the project instruction file a coding agent reads before touching your repo — CLAUDE.md, AGENTS.md or equivalent. Insists every command in the file be executed before it is written down, and that the file record observed agent failures and hard constraints rather than restating what the code already says. Triggers when you ask to write, refresh or fix a project instruction file.

How to install Agent Instructions

git clone https://github.com/nimadorostkar/Claude-Skills-collection.git
mkdir -p ~/.claude/skills
cd Claude-Skills-collection && cp -r skills/agent-tooling/agent-instructions ~/.claude/skills/agent-instructions

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Agent Instructions

/agent-instructions Writes a CLAUDE.md/AGENTS.md that lists only verified commands and real gotchas

It also activates on plain-language prompts like these:

Write a CLAUDE.md for this repo listing only verified commands
Refresh our AGENTS.md so it reflects real agent failures we hit
Fix our project instruction file, it has commands that don't run

Frequently asked questions

Is the Agent Instructions skill free?: Yes. The skill itself is free from nimadorostkar/Claude-Skills-collection. SkillProof publishes the install command and an independent test verdict at no cost.
Does Agent Instructions work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Aug 7, 2026. Verdict: Tested · Works. Wrote a CLAUDE.md for the consensus-loop repo twice and then executed every command in both files. The no-skill version shipped five commands of which three were broken, including `node --test tests/` — the exact command the repo's real CLAUDE.md carries, which dies on Node 26 with 'Cannot find module <repo>/tests'. Following the skill's rule that commands must be run before they are written turned two of those traps into explicit do-not-run lines with the real error text, and surfaced a third: `npm run quality` in test-harness cannot pass because @types/node is missing, so `npx vitest run` (44/44 green) replaced it. Cost of the better file was 59 lines against 39.
What is the Agent Instructions SkillProof Score?: 9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.
How do I install Agent Instructions?: Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Agent Instructions with Cursor, Copilot, Gemini CLI, Codex or other AI tools?: The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.