Why skillpack?
AI coding assistants (Claude Code, Cursor, Codex, OpenCode, GitHub Copilot, Freebuff) interact with software differently than humans:
- Humans read high-level READMEs, tutorials, and manual pages to learn syntax.
- AI Agents look for machine-readable skill files, plugin manifests, rule definitions, and structured invocation syntax.
Without an agent guidance layer, AI coding agents frequently:
- Miss installed tools and attempt to write redundant custom scripts from scratch.
- Hallucinate invalid CLI flags, resulting in execution failures and broken scripts.
- Waste context tokens and time repeatedly grepping
--helpoutputs across multiple exploratory reasoning steps.
skillpack solves this with a single command. It introspects your project repository, discovers your CLI binaries or library interfaces, extracts verified argument signatures, and generates tailored distribution files for all major agent ecosystems.
Key Features
- Universal Multi-Ecosystem Generation: Generates native guidance files for 14 agent formats simultaneously (
AGENTS.md, Claude Code (plugin + native.claude/skills/), Cursor.mdc, Codex, OpenCode, Copilot,CLAUDE.md,GEMINI.md, Windsurf, Aider, Cline, Roo Code, Kilo Code, and Goose). Freebuff and other AGENTS.md-native agents read the generatedAGENTS.md. - Zero-Drift Verification: Simulates agent invocations against your live CLI to verify that every documented flag and subcommand actually exists.
- Pre-Commit Safe:
skillpack initvalidates the complete surface before writing a single file to disk. - Non-Destructive Updates:
skillpack updaterefreshes flags and versions while preserving your hand-written descriptions and notes. - Turnkey CI Integration: Includes GitHub Actions workflows and
.pre-commithooks to prevent documentation drift in pull requests.
Quick Start
1. Installation
# One-command installer (prebuilt binary, no Rust toolchain)
|
# Or via Homebrew
# Or via cargo-binstall (prebuilt binary)
&&
# Or build from source
2. Scaffold Agent Guidance
Run interactive setup in your project root:
# Interactive setup (asks a few quick questions and persists answers to skillpack.toml)
# Or generate for all supported ecosystems at once:
3. Zero-Interaction Automated Setup (--auto)
For instant, unattended generation in CI or local scripts:
# Introspects README, git config, LICENSE, and compiled CLI binaries automatically
4. Verify Accuracy
Validate your repository's agent guidance against your real CLI:
What skillpack Generates
Running skillpack init --target all generates a clean, non-intrusive distribution layer:
| Target / Ecosystem | Generated File(s) | Description |
|---|---|---|
| AGENTS.md | AGENTS.md |
Standard instructions file read natively by Freebuff, Zed, Cursor, Windsurf, and 60k+ repos. |
| Claude Code | .claude-plugin/, skills/<tool>/SKILL.md & .claude/skills/<tool>/SKILL.md |
Plugin manifest + skill spec, plus the native .claude/skills/ directory Claude auto-loads with no install step. |
| Codex | .codex/skills/<tool>/SKILL.md |
Skill specification for the OpenAI Codex CLI (same SKILL.md shape as Claude Code). |
| Cursor | .cursor/rules/<tool>.mdc |
Context-aware rule file with automated file-glob matching. |
| OpenCode | .opencode/agents/<tool>.md |
Agent definition for OpenCode AI coding environments. |
| GitHub Copilot | .github/copilot-instructions.md |
Custom repository instructions for GitHub Copilot. |
| Gemini CLI | GEMINI.md |
Native repository instruction layer for Google Gemini CLI. |
| CLAUDE.md | CLAUDE.md |
Root instructions file read by Claude Code, Cline, and Roo Code. |
| Windsurf | .windsurf/rules/<tool>.md |
Cascade IDE rule file. |
| Aider | CONVENTIONS.md |
Codebase convention guidelines for Aider. |
| Cline | .clinerules/<tool>.md |
Workspace rules for the Cline coding agent (optional paths: conditional frontmatter). |
| Roo Code | .roo/rules/<tool>.md |
Workspace rules for Roo Code. |
| Kilo Code | .kilocode/rules/<tool>.md |
Workspace rules for Kilo Code (auto-included directory). |
| Goose | .goose/instructions.md |
Project instructions for Block's Goose agent. |
| Deterministic Config | skillpack.toml |
Committed configuration making future updates and CI checks deterministic (JSON Schema at skillpack.schema.json). |
Measured Benchmark: The skillpack Impact
We benchmarked autonomous coding agents (Google Antigravity agy with Gemini 3.7 Flash) performing complex search and execution tasks on a plain repository clone versus a skillpack-guided clone:
Baseline (Plain Clone): [Help Grep ×4] ──> [Wandered into other projects] ──> [Success in 60 Rounds] (51s)
Guided (With skillpack): [Verified Flags, Anchored in Repo] ───────────────────> [Success in 40 Rounds] (35s)
Performance Metrics (Google Antigravity / sharkdp/fd Benchmark Suite, 3 runs per condition)
| Metric | Plain Clone (Baseline) | Clone + skillpack | Improvement |
|---|---|---|---|
| Wall Clock Latency | 50.8 s | 34.5 s | 32% Faster |
| Reasoning Rounds | 60.0 steps | 40.0 steps | 33% Fewer Steps |
| Help Query Detours | 4.0 queries | 2.0 queries | 50% Fewer |
| Token Consumption | 146.1k tokens | 100.5k tokens | 31% Fewer |
| Task Accuracy | 4.0 / 4 (median) | 4.0 / 4 (every run) | Guided perfect in all runs |
Full benchmark methodology, repeatable test suites (
fd,ripgrep,bat), and replay harness documentation are available indocs/benchmark.md. Want to see it happen? Read the step-by-stepdocs/agent-demo.md, or reproduce the A/B run yourself from the committed transcripts.
Command-Line Reference
skillpack init: Scaffold agent distribution files (interactive,--auto, or--non-interactive).skillpack verify: Check guidance files against agent schemas and live--helpflag surfaces (supports--min-score <N>,--fix, and--format {human,json,sarif,github,junit}).skillpack doctor: Diagnose language and CLI candidate discovery decision traces.skillpack update: Incrementally regenerate files fromskillpack.tomlafter CLI changes.skillpack diff: Preview pending guidance updates without modifying disk.
CI/CD & Pre-Commit Integration
GitHub Actions Workflow
Use the reusable workflow to prevent documentation drift in PRs (it installs a
pinned skillpack and runs verify across the OS matrix):
name: Verify Agent Guidance
on:
push:
branches:
pull_request:
branches:
jobs:
verify:
uses: nordicnode/skillpack/.github/workflows/skillpack.yml@v0.13.2
For a stricter gate that fails on any warning (not just critical failures),
call skillpack verify --min-score 100 directly — that is the bar skillpack
holds itself to in its own CI.
Pre-Commit Hook
Add to your .pre-commit-config.yaml:
repos:
- repo: https://github.com/nordicnode/skillpack
rev: v0.13.2
hooks:
- id: skillpack-verify
Supported Language Ecosystems
skillpack automatically detects CLI binaries and library entrypoints across:
- Rust: Built artifacts in
target/{release,debug}(including Cargo workspaces) and PATH binaries. - Node.js:
package.jsonbinfields and./bin/<name>.jsscripts. - Python:
pyproject.toml[project.scripts]and Poetry/Flit entries. - Go:
go run .forpackage mainmodules and built binaries. - Ruby: Gem binstubs in
exe/andbin/. - PHP: Composer package binaries in
vendor/bin. - JVM: Gradle
installDistand Maven shaded JAR entrypoints. - .NET / C#: SDK-style executable
.csprojtargets. - Zig:
build.zig/build.zig.zonand artifacts inzig-out/bin/. - Swift:
Package.swiftand.build/{debug,release}/artifacts orswift run. - C / C++:
CMakeLists.txt,meson.build,Makefileandbuild//bin/binaries. - Elixir:
mix.exsand_build/{dev,prod}/rel/releases ormix escript. - Deno:
deno.json/deno.jsoncanddeno runscript entrypoints.
License & Community
skillpack is open-source software licensed under the MIT License.
- Documentation: See
docs/reference.mdfor full check catalogs and schemas. - Changelog: See
CHANGELOG.mdfor version release notes. - Contributing: Contributions are welcome! Review
CONTRIBUTING.mdto get started.