OpenCode Planning Toolkit

Keep your projects organized and your AI agents aligned.

This plugin adds structure to your development workflow with reusable specifications, actionable plans, and progress tracking—all integrated directly into OpenCode.

Why This Plugin

• Enables planning with any and all agents
• Organize work into reusable specs and actionable plans
• Track progress with clear status (active/done)
• Maintain project-wide standards that apply across all work
• Automatically receive <available_plans> in the system prompt with all plan names and descriptions
• Bundled skill automatically loads to guide agents through proper planning workflow

Tools

• create_spec - Create a reusable specification (repo-level or feature-specific)
• create_plan - Create an actionable work plan with implementation steps (min 5)
• append_spec - Link an existing spec to a plan
• read_plan - Read a plan with all linked spec content expanded inline
• mark_plan_done - Mark a plan as complete

Bundled Skill

This plugin includes a bundled skill (plans-and-specs) that automatically loads with the plugin. The skill provides agents with:

• Clear workflow instructions for creating plans and specs
• Proper order of operations (create plan → append REPO specs → ask about FEATURE specs)
• Instructions to check for existing plans before creating new ones
• Guidance on when to use each tool

When agents use it:

• User asks to create a plan, roadmap, or break down work
• User mentions specs, requirements, or standards documentation
• User references existing plans that need reading or updating
• User asks to mark work complete or link requirements

The skill ensures agents follow best practices automatically without manual prompting.

Installation

Add to your opencode.json:

{
  "plugins": [
    "@howaboua/opencode-planning-toolkit@latest"
  ]
}

Or for local development:

{
  "plugins": [
    "file:///path/to/opencode-planning-toolkit/index.ts"
  ]
}

Quick Start

1. Create a Reusable Spec

Define standards or requirements once, use them everywhere:

Ask the agent: "Create a repo-level spec for TypeScript coding standards"

The agent creates docs/specs/typescript-standards.md:

# Spec: typescript-standards

Scope: repo

- Use strict mode
- All functions must have explicit return types
- Prefer `const` over `let`
- No `any` types without justification

2. Create a Plan

Turn a goal into actionable steps:

Ask the agent: "Create a plan for adding user authentication"

The agent creates docs/plans/user-auth.md:

---
plan name: user-auth
plan description: JWT authentication for API
plan status: active
---

## Idea
Add secure JWT authentication to the API with login, logout, and token refresh.

## Implementation
- Design JWT token structure and expiry policy
- Add /auth/login endpoint with password validation
- Add /auth/refresh endpoint for token renewal
- Add /auth/logout endpoint to invalidate tokens
- Write tests for all auth endpoints

## Required Specs
<!-- SPECS_START -->
<!-- SPECS_END -->

Agent automatically:

1. Links any existing REPO scope specs
2. Asks if you want a FEATURE spec for this plan

3. Link Specs to Plans

Attach relevant specs to ensure standards are followed:

Ask the agent: "Link typescript-standards spec to the user-auth plan"

The plan updates:

## Required Specs
<!-- SPECS_START -->
- typescript-standards
<!-- SPECS_END -->

4. Read Before Work

When starting implementation, the agent reads the full context:

Ask the agent: "Read the user-auth plan and start implementing"

The agent receives the full plan with all linked spec content expanded inline.

5. Mark Complete

Ask the agent: "Mark the user-auth plan as done"

Status updates from active → done.

File Organization

your-project/
├── docs/
│   ├── specs/           # Reusable specifications
│   │   └── *.md         # Each spec is a markdown file
│   └── plans/           # Work plans
│       └── *.md         # Each plan is a markdown file

Specs naming: {name}.md (e.g., typescript-standards.md) Plans naming: {name}.md (e.g., user-auth.md)

License

MIT