Is the README Discipline skill free?

Yes. The skill itself is free from Skillproofdev/readme-discipline. SkillProof publishes the install command and an independent test verdict at no cost.

What is the README Discipline SkillProof Score?

8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 5/5.

How do I install README Discipline?

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 README Discipline 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.

README Discipline

Code-grounded READMEs: every claim traced to file:line, every command verified to run.

by SkillProof · Skillproofdev/readme-discipline

Tested · Works

Install — copy the command Download source ↗

Test report

Verdict: Tested · Works
Score: 8.8/10
Tested: Jul 11, 2026
Environment: Claude Code 2.x (agent harness)
Upstream re-checked: Aug 10, 2026 · 3f8f089

Our own skill — tested under the same protocol, disclosed. Bench measured 3 frozen OSS repos, baseline vs skill agent: skill hit 18/18 section completeness vs 14/18, was preferred 3/3 with trust 4.67 vs 3.67, and had fewer fabricated claims (1 vs 2, a narrow margin since both baselines borrowed accurate content from existing docs). Caveats disclosed in the verdict: single expert rater (not the ≥3-developer panel the protocol calls for), N=3, and the skill logged its own fabrication (an HTTP 400/500 mix-up) rather than hiding it. Protocol test caught real issues in our own repos (dead benchmark links, strict-YAML frontmatter) — fixed and re-verified same day.

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 7/10
Docs & honesty 5/5

What README Discipline does

Forces README generation and audits to trace every claim to source before it can be written — features must be found in code, commands must be copied from real manifests and run, and a sourced Limitations section is mandatory. Benchmarked against baseline agents on 3 real OSS repos with a clean win on completeness and blind preference, though the fabrication-reduction margin is narrow and the evaluation used a single rater.

README Discipline before / after — same task, with and without

Without the skill

Baseline README for python-shaarli-client states "Requirements: Python 3.4+" and documents `tox -e py34 -e py36` — neither environment exists in the project's actual tox.ini (which only supports 3.10-3.14); both commands fail live with "could not find python interpreter with spec py34". Two fabricated claims traced to stale prose, not source.

With README Discipline

Skill-generated README for the same repo removes the false Python 3.4+ claim, documents only the real tox environments found in tox.ini, and ships a sourced Limitations section. Result: 0 fabrications, 7/7 documented commands executable, preferred by the developer rater over the baseline with higher trust (5 vs 3).

How to install README Discipline

git clone https://github.com/Skillproofdev/readme-discipline ~/.claude/skills/readme-discipline

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

Commands — how to trigger README Discipline

/readme-discipline Code-grounded READMEs: every claim traced to file:line, every command verified to run.

It also activates on plain-language prompts like these:

Write a README for this repo
Can you review our README.md and flag anything that's inaccurate or outdated?
I'm scaffolding a new npm package — generate the initial README along with package.json

Frequently asked questions

Is the README Discipline skill free?: Yes. The skill itself is free from Skillproofdev/readme-discipline. SkillProof publishes the install command and an independent test verdict at no cost.
Does README Discipline work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Jul 11, 2026. Verdict: Tested · Works. Our own skill — tested under the same protocol, disclosed. Bench measured 3 frozen OSS repos, baseline vs skill agent: skill hit 18/18 section completeness vs 14/18, was preferred 3/3 with trust 4.67 vs 3.67, and had fewer fabricated claims (1 vs 2, a narrow margin since both baselines borrowed accurate content from existing docs). Caveats disclosed in the verdict: single expert rater (not the ≥3-developer panel the protocol calls for), N=3, and the skill logged its own fabrication (an HTTP 400/500 mix-up) rather than hiding it. Protocol test caught real issues in our own repos (dead benchmark links, strict-YAML frontmatter) — fixed and re-verified same day.
What is the README Discipline SkillProof Score?: 8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 5/5.
How do I install README Discipline?: 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 README Discipline 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.