Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/modiqo/skillspec/llms.txt

Use this file to discover all available pages before exploring further.

SkillSpec sits alongside your existing SKILL.md files and adds structure where prose alone falls short — routes, rules, dependencies, tests, and a proof record you can read after every run. No new runtime, no orchestration platform; just a CLI and a small skill.spec.yml that lives next to your skill.

Introduction

Understand what SkillSpec is, what problems it solves, and how it fits into your agent workflow.

Quickstart

Run your first Doctor report and port a skill to a SkillSpec contract in under five minutes.

Installation

Install the CLI via the release installer, Cargo, or your agent harness plugin marketplace.

Command Reference

Browse every skillspec command — runtime, authoring, router, and installation.

​
The Assess → Port → Prove Loop

SkillSpec is built around three actions you repeat for every skill you care about.
1

Assess with Doctor

Point Doctor at any skill folder or public GitHub URL. You get an agent follow-through risk score, findings about buried rules, undeclared tools, and missing proof — before you touch any code.
skillspec doctor ./my-skill
2

Port to a Contract

Import the prose skill into a typed skill.spec.yml with routes, rules, forbidden actions, dependencies, checks, and tests. Compile it to a thin harness loader.
skillspec port-one-shot ./my-skill --out ./my-skill-spec --target codex-skill --prove
3

Prove Every Run

Each run produces a decision trace and execution ledger. Align them against the contract to get a proof record: selected route, completed steps, missing evidence, and token metrics.
skillspec trace align ./skill.spec.yml \
  --decision-trace .skillspec/traces/run-123 \
  --execution-trace .skillspec/traces/run-123/execution.jsonl \
  --quiet

​
Core Concepts

Skill Contracts

Learn how skill.spec.yml structures routes, rules, elicitations, and proof alongside your prose SKILL.md.

Doctor Risk Reports

Understand how Doctor scores follow-through risk, detection methods, and what each finding means.

Routes & Rules

Explore the decision algebra: how routes are selected, how rules compose, and how elicitations collect missing facts.

Traces & Alignment

See how append-only traces and execution ledgers produce verifiable alignment reports.

​
Guides

Assess a Skill

Run Doctor locally or via the hosted page to get a baseline risk report for any skill.

Import & Port

Walk through the full import workflow from source map to validated, compiled contract.

Router Mode

Install the skill router to handle large skill libraries and control harness-native visibility.

Workspace Skills

Map, import, compile, and install multi-skill or plugin-shaped workspace repositories.

Build docs developers (and LLMs) love

Get started for freeTalk to us