Skill Conductor

Install

npx skills add smixs/skill-conductor

This is third-party code. Review the repository files before installing.

What it does

Covers the whole skill lifecycle in six modes: create, improve, validate, review third-party skills, optimize the description and package as a .skill file. Before any SKILL.md is written it makes you pick one of five architecture patterns and confirm the agent actually fails without the skill. Quality is scored with BinEval: binary questions backed by evidence and a gate on critical questions. It ships grader, comparator, analyzer and bineval agents plus Python scripts for running evals.

Who it is for. Authors of skills for Claude Code and compatible agents who want measurable quality checks instead of eyeballing.

Example request

Build a skill that teaches the agent to write changelogs our way, then run evals on it

Limitations

Script-based modes require uv and LLM access; without uv the skill stops. Evals and improvement loops consume a noticeable amount of tokens. Cross-family judge calibration assumes access to a second model.

How to disable. Remove the plugin via /plugin in Claude Code or delete the skill-conductor folder from ~/.claude/skills.

Security check

• Runs Python scripts via uv
• Evals call LLMs and consume tokens

README in short

Skill Conductor presents itself as a meta skill for the full skill lifecycle: design, build, test, evaluate and package. Its method combines Anthropic's skill-creator, the writing-skills guide from Superpowers and several papers on LLM evaluation. Version 3 moved to BinEval scoring and a gated self-update loop with a held-out split. It installs via skills.sh or as a Claude Code plugin. Licensed under MIT.

SKILL.md

---
name: skill-conductor
description: >
  Create, edit, evaluate, and package agent skills. Use when building a new
  skill from scratch, improving an existing skill, fixing a skill that never
  triggers or fires unreliably, running evals to test a skill, benchmarking
  skill performance, optimizing a skill's description, reviewing third-party
  skills for quality, or packaging skills for distribution — even if the user
  doesn't explicitly say "skill" (e.g. "teach Claude to do X", "make the
  agent always follow Y"). Not for using skills or general coding
  tasks.
---

# Skill Conductor

Full lifecycle management for agent skills: **draft → test → review → improve → repeat**.

One skill to rule them all — from architecture to packaging. The core loop is always the same: write something, test it, see what fails, fix it, test again.

## Runtime requirements (pre-flight)

Before any mode that touches scripts (CREATE, IMPROVE, VALIDATE, OPTIMIZE, PACKAGE), run the pre-flight block → **`references/runtime-setup.md`** (checks `uv`, sets `UV_BIN`/`SKILL_CONDUCTOR_DIR`, verifies LLM access). If `uv` is absent, stop and tell the user.

## How to communicate

Read context cues. If the user is a skill author iterating on their own work, be direct and technical. If they're new to skills, explain the _why_ behind each step — not just what to do, but why it matters. Default to conversational, not robotic.

- Explain trade-offs when there's a real choice to make
- Use concrete examples over abstract rules
- When something fails, explain the root cause, not just the fix
- Imperative voice in instructions: "Extract the data", not "You should extract"

## Modes

Detect mode from context. If ambiguous, ask.

| Mode        | When                                             | What happens                                                    |
| ----------- | ------------------------------------------------ | --------------------------------------------------------------- |
| 1. CREATE   | "build a skill", "new skill for..."              | Full lifecycle: intent → architecture → scaffold → write → test |
| 2. IMPROVE  | "fix this skill", "it doesn't trigger"           | Diagnose → eval loop → gated self-update → iterate              |
| 3. VALIDATE | "test this skill", "run evals"                   | Structural checks + trigger testing + BinEval scoring           |
| 4. REVIEW   | "review this skill", third-party assessment      | 11-point quality gate, quick and focused                        |
| 5. OPTIMIZE | "improve triggering", "description optimization" | Automated description optimization with train/test split        |
| 6. PACKAGE  | "package for distribution"                       | Validate + bundle into .skill file                              |

Related