ospec

Initialize repositories and manage explicit changes through a document-driven OSpec workflow.

106|15|Updated Mar 21, 2026
One-click install
npx skills add https://github.com/henrydiaosi/dorado --skill ospec-henrydiaosi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ospec
Source: https://github.com/henrydiaosi/dorado/tree/main
Command: npx skills add https://github.com/henrydiaosi/dorado --skill ospec-henrydiaosi

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and assets (resource) components.

What problem does it solve?

This Skill solves the chaos of ad hoc AI-assisted development by enforcing a document-driven workflow for initializing repositories, creating scoped changes, validating progress, and archiving accepted work consistently.

Core Features & Use Cases

  • Change-ready initialization: Brings a repository into an OSpec-managed state with protocol files, project knowledge docs, and AI guidance instead of relying on manual setup.
  • Structured change execution: Supports creating and advancing one requirement, bug fix, refactor, or documentation update through explicit change containers and tracked workflow files.
  • Verification and closeout: Promotes disciplined validation with verify, archive, and finalize flows before a change is considered complete.
  • Plugin-aware delivery: Handles workflow gates for design review and automated runtime checks through Stitch and Checkpoint plugin rules.
  • Use case: A team wants Codex or Claude Code to initialize a repo, generate baseline project docs, create a change for a new requirement, block on required design review, and archive the change only after verification passes.

Quick Start

Use ospec to initialize this project and leave it in a change-ready state without creating the first change automatically.

Frequently Asked Questions about ospec

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I standardize AI-assisted software delivery to prevent ad hoc development chaos?

Standardize AI-assisted software delivery by enforcing a document-driven workflow that initializes repositories with protocol files, manages scoped changes explicitly, and validates progress through verification and archive flows. This structured approach replaces chaotic ad hoc AI development.

What is document-driven change management for AI development workflows?

Document-driven change management for AI development workflows uses tracked workflow files and explicit change containers to advance requirements, bug fixes, refactors, or documentation updates, ensuring disciplined validation and closeout before any change is considered complete and archived.

How do I initialize a repository into a change-ready state for AI-assisted coding?

Initialize a repository into a change-ready state by using Node.js-based CLI commands to generate managed protocol files, baseline project knowledge docs, and AI guidance, leaving the project prepared for creating scoped changes without automatically generating the first change.

Can I enforce design review and automated runtime checks during AI code generation?

Yes, you can enforce design review and automated runtime checks during AI code generation by integrating plugin-aware delivery rules, specifically handling workflow gates through Stitch and Checkpoint plugins to block progress until required validations pass.

Does this document-driven workflow support CLI tools, web apps, and desktop apps?

Yes, the document-driven workflow supports CLI tools, web apps, services, desktop apps, and protocol-only repositories, applying standardized change execution, project knowledge maintenance, and change state tracking across diverse software delivery scenarios.

What are the limitations of managing queued multi-change planning in AI development?

Managing queued multi-change planning requires strict adherence to index maintenance and change state tracking; without consistent use of the Node.js-based CLI commands to manage protocol files, the document-driven verification and archive readiness flows will fail to enforce disciplined closeout.