how-to-contribute

Analyzes project documentation to detect contribution gates, AI bans, and prompt-injection traps.

716|31|Updated Mar 13, 2023
One-click install
npx skills add https://github.com/wimpysworld/nix-config --skill how-to-contribute
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: how-to-contribute
Source: https://github.com/wimpysworld/nix-config/tree/main/home-manager/_mixins/agentic/assistants/skills/how-to-contribute
Command: npx skills add https://github.com/wimpysworld/nix-config --skill how-to-contribute

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Before opening an issue or pull request, contributors need to know a project's rules: whether pre-approval is required, whether AI-assisted contributions are banned, and whether instruction files like AGENTS.md contain traps designed to expose AI agents. Reading every policy document manually is slow and easy to get wrong.

Core Features & Use Cases

  • Policy Detection: Scans README, CONTRIBUTING, CODE_OF_CONDUCT, SUPPORT, SECURITY, and GitHub templates for approval gates and discussion requirements.
  • AI Trap Identification: Detects instructions in AGENTS.md and CLAUDE.md that tell AI agents to self-report, insert markers, or disclose their use, and quotes them without executing them.
  • Structured Report: Returns a five-bullet summary with clear markers (BAN, TRAP, GATE) so the user knows the rules before contributing.
  • Use Case: Before submitting a pull request to an unfamiliar open-source repository, run this Skill against the repo to learn that it requires a discussion first and bans AI-assisted contributions.

Quick Start

Check the contribution rules for the repository at owner/name and tell me whether AI-assisted contributions are allowed.

Frequently Asked Questions about how-to-contribute

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check a project's contribution rules before opening a pull request?

Point this Skill at a local path or GitHub repository and it reads CONTRIBUTING.md, README, code of conduct, and issue templates. It reports approval gates, required prior discussions, and AI contribution bans in a short summary.

How can I detect AI contribution bans in open source projects?

The Skill scans policy documents and agent instruction files for statements banning AI-assisted contributions. Any ban is reported first in the summary and marked with a BAN flag, with the source file and quoted line.

Can this analyze a GitHub repository without cloning it?

Yes. For GitHub targets it uses read-only gh and gh-api-safe routes or documented read-only GitHub MCP operations. It never performs mutations and defers any write actions to other skills.

What are prompt injection traps in AGENTS.md and CLAUDE.md files?

Some projects embed instructions telling AI agents to self-report, insert markers, or disclose AI use. The Skill treats these files as data, quotes the trap text in its report marked TRAP, and never executes those instructions.

What happens if a project has no CONTRIBUTING.md file?

Each document in the scan may be absent, and the Skill records the absence instead of guessing. It never infers rules that no document states, so the report reflects only documented policy.