Rules and skills that teach AI agents how to contribute to open source projects without being the villain.
82
90%
Does it follow best practices?
Impact
78%
5.20xAverage score across 7 eval scenarios
Critical
Do not install without reviewing
Rules, skills, and scripts that teach AI agents how to contribute to open source projects without being the villain.
tessl install tessl-labs/good-oss-citizenInstall on your fork of whatever OSS project you're contributing to. The tile loads only in that project's context — switch to your own codebase and it isn't there.
AI agents write working code but ignore everything around it: contribution guidelines, AI policies, prior rejected PRs, claimed issues, DCO requirements, changelog updates. This tile teaches the agent to check all of that before submitting.
Without the tile: agents score ~22% on OSS contribution process checks (while writing perfectly functional code). With the tile: 99%.
The tile has three layers:
Every rule and skill instruction traces back to documented failure modes from real OSS projects. See RESEARCH.md for all 16 findings with sources.
You don't have to ban AI — you can guide it instead. Add a tessl.json to your repository root:
{
"name": "your-project",
"dependencies": {
"tessl-labs/good-oss-citizen": {
"version": "1.0.1"
}
}
}Then mention it in your README, CONTRIBUTING.md, or AI_POLICY.md:
We accept considerate AI-assisted contributions. Install
tessl install tessl-labs/good-oss-citizenin your fork first.
AI agents read contribution guidelines before they start working. When an agent sees this instruction, it installs the tile and invokes it automatically — no manual setup from the contributor required.