Back to directory

MapLibre Agent Skills

A community-maintained collection of markdown skills that give AI coding assistants guidance for writing correct MapLibre GL JS mapping code.

MapLibre Agent Skills

MapLibre Agent Skills is a set of markdown files that AI coding assistants read as context when implementing MapLibre applications, giving the AI judgment to avoid common API mistakes and follow patterns that work. New skills are prioritized by reported AI failures, recent API changes, and long-standing issues from GitHub, Stack Overflow and the community Slack.

Each skill aims to be accurate to the MapLibre APIs, actionable rather than only descriptive, and warranted by a real failure an AI assistant would otherwise make. Skills are tested with Promptfoo evaluations against real developer questions, and a skill's status, verified or provisional, records whether an evaluation has proven it fixes a real failure.

It is a JavaScript-based repository of skill files rather than a runnable application.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.