Back to directory

Distilly

A tool that distills a person's experience, judgment, voice and working style, from messages, documents and interviews, into a reusable profile packaged as an Agent Skill for AI agents and bots.

Distilly

Distilly, formerly named Colleague Skill, is a person-modeling layer for AI agents. It turns source material such as messages, documents, interviews and public sources, together with a description from the user, into a portable, source-grounded Person Profile built from observable experience, decision patterns, expression and ways of working. The project states that a profile does not claim to clone the person behind it.

It is meant to capture the working style of a colleague, mentor or teammate, or the voice of a family member, friend, public figure or fictional character, so an agent can draw on that grounded profile. The current release packages each profile as an Agent Skill, so any host that supports Agent Skills can install and invoke it; the canonical skill is named distilly.

It is written in Python and supports local skill discovery for hosts such as Claude Code, Hermes, OpenClaw, Codex, DeepSeek Harness, Pi, Grok Build and OpenCode.

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.