Install
$ agentstack add skill-pradnyeshp-claude-skills-contributing ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
Contribution signals
!ls CONTRIBUTING* CODE_OF_CONDUCT* .github 2>/dev/null && echo "(existing files above — update, don't blindly overwrite)"
!cat package.json pyproject.toml Cargo.toml Makefile 2>/dev/null | grep -iE '"scripts"|"test"|"lint"|"format"|test =|lint =|^[a-z-]+:' | head -25
!ls .pre-commit-config.yaml .husky .github/workflows .github/PULL_REQUEST_TEMPLATE* 2>/dev/null
Instructions
Generate or update CONTRIBUTING.md for this project (focus on the $ARGUMENTS section if one was named). Write for a new contributor who wants to make their first change correctly.
Method
- Read the project's real workflow from the signals above — the dev/build/test/lint commands (scripts, Makefile, tox), the hooks (pre-commit/husky), CI (what the workflows enforce), and any PR template or existing conventions. Document what the project actually does, not a generic template.
- Decide create vs update. If
CONTRIBUTING.mdexists, preserve its structure and any project-specific rules; refresh only what's stale or missing. Otherwise create one from the structure below. - Write
CONTRIBUTING.mdat the repo root (or.github/), and don't duplicate the README — link to it for setup rather than repeating it.
Suggested structure
# Contributing
## Getting set up # prerequisites + the actual setup commands (link README if covered)
## Development workflow # branch naming, how to run the app, where code lives
## Tests & checks # the exact commands CI runs — tests, lint, typecheck, format
## Commit & PR guidelines # commit convention, PR size/scope, what a good PR looks like
## Reporting issues # how to file a bug / request a feature
## Code of conduct # link, if a CODE_OF_CONDUCT file exists
Rules
- Document the real, verifiable process — every command must come from a script/Makefile/CI target that exists. If you can't confirm something (e.g. the branching model), mark it as a suggested default rather than asserting it.
- Reflect what CI actually enforces so contributors can pass checks locally before pushing — point them at the
pre-commit/lint-fix/test-genskills where relevant. - If the repo uses Conventional Commits (check git history / commitlint config), state that and link the
commit-messageskill; don't impose a convention the project doesn't use. - Keep it practical and skimmable — a new contributor should be able to go from clone to passing PR by following it. Don't pad with boilerplate that doesn't apply.
- After writing, summarize what you created or changed and flag anything you marked as a suggested default needing the maintainer's confirmation.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Pradnyeshp
- Source: Pradnyeshp/Claude-Skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.