Back to directory

OnurPi

A personal but reusable collection of extensions, tools, skills, a theme and settings for the Pi coding agent, packaged as separate installable npm packages that others can adopt individually.

OnurPi

OnurPi packages the author's day-to-day setup for the Pi coding agent as a set of self-contained npm packages, so pieces can be installed or adapted individually rather than adopting a single monolithic configuration.

Packages cover agent behavior and reliability, including bounded transcript folding, opt-in loop detection and interruption, infinite capped retries with a manual wake key, and a stable transport policy for context compaction; interface conveniences such as a live status line, a spinner, an editable prompt queue and an animated context display; account and usage management, including account-profile switching with fallback routing and multi-provider usage reports; and safety tooling such as fail-closed checks for destructive shell commands and persistent, race-free shell sessions.

Other packages add a portable theme, skill-slug invocation, branch-aware task planning with live progress, and a read-only local-model issue-triage workflow. It is written in TypeScript.

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.