MCPInstall

idea-to-plan

Turn a vague idea or feature request into a written, reviewable spec and implementation plan before any code is written. Use when the user says "I want to build X", "help me think through", "I have an idea", or hands over an underspecified request.

Installs
56
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/composio-community/opencode-skills --skill idea-to-plan
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads idea-to-plan automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

Idea to Plan

Convert an underspecified request into (1) a one-paragraph problem statement, (2) a scoped spec, and (3) a phased implementation plan — each shown to the user for sign-off before moving on. Never jump to code from a vague prompt.

When to use

  • The request is a goal, not a specification ("build a dashboard", "add auth").
  • Requirements, constraints, or success criteria are missing or implied.
  • The user is exploring whether something is worth building.

Do not use this for well-specified tasks, bug fixes with a clear repro, or mechanical changes — those go straight to implementation.

Procedure

1. Restate the problem (before asking anything)

Write one paragraph: who has what problem, why now, and what "solved" looks like. Show it. If you can't write it, you don't understand the request — that gap is your first question.

2. Ask only decision-changing questions

Batch 3–6 questions that would change the design if answered differently. Skip anything with a safe default — state the default instead and move on. Good targets:

  • Scope boundaries (what is explicitly out)
  • Hard constraints (stack, deadline, data, compliance)
  • Success criteria (how we'll know it works)
  • The one assumption that, if wrong, wastes the most work

3. Write the spec

## Problem
## Goals / Non-goals
## Constraints
## Approach (1–3 sentences)
## Open questions

Keep it under one screen. Show it, get explicit "yes" on Goals/Non-goals before planning.

4. Write the phased plan

Break into phases that each deliver something verifiable:

Phase 1: <smallest end-to-end slice that proves the approach> — verify by <X>
Phase 2: ...

Each phase: concrete deliverable, the files/areas touched, and how it's verified. Flag dependencies and the riskiest phase. Front-load risk.

5. Hand off

Output the spec + plan as the final artifact. Recommend the first phase and stop — implementation is a separate step the user triggers.

Quality bar

  • Problem statement fits in one paragraph and names a real user.
  • Non-goals are explicit (a spec with no non-goals is not scoped).
  • Every phase is independently verifiable.
  • No question asked that has an obvious default.
  • Zero code until the spec is signed off.

Anti-patterns

  • Asking 15 questions when 4 decide the design.
  • "Phase 1: set up the project" with no verifiable outcome.
  • Restating the user's prompt back as if it were a spec.
  • Designing the gold-plated version before the wedge is agreed.

FAQ

What does the idea-to-plan skill do?

Turn a vague idea or feature request into a written, reviewable spec and implementation plan before any code is written. Use when the user says "I want to build X", "help me think through", "I have an idea", or hands over an underspecified request.

How do I install idea-to-plan?

Run npx skills add https://github.com/composio-community/opencode-skills --skill idea-to-plan in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/idea-to-plan/ (personal) or .claude/skills/idea-to-plan/ (project).

Which AI agents can use idea-to-plan?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is idea-to-plan safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains idea-to-plan?

It is published by composio-community in composio-community/opencode-skills. It has 56 recorded installs.

More from composio-community/opencode-skills

View all →

frontend-design-systems

composio-community/opencode-skills

Open the skill to read its SKILL.md instructions.

Frontend & Design140 installs

playwright-mcp-workflows

composio-community/opencode-skills

Workflow for using Playwright with MCP-based agent systems to automate browsers, test applications, perform UI validation, and execute reliable end-to-end workflows through structured browser orchestration.

Testing & QA96 installs

vercel-react-best-practices

composio-community/opencode-skills

Workflow for building production-quality React applications using Vercel-style best practices focused on readability, performance, maintainability, and stable UI behavior.

repo-context-packaging

composio-community/opencode-skills

Open the skill to read its SKILL.md instructions.

70 installs

multi-agent-orchestration

composio-community/opencode-skills

Split a large task across parallel sub-agents with clean role boundaries and a merge step. Use when the user asks to "parallelize", "use multiple agents", "fan out", or when a task has independent workstreams that bottleneck on one agent.

context-engineering

composio-community/opencode-skills

Workflow for managing, structuring, compressing, and orchestrating context effectively for Claude and AI agent systems to improve reasoning quality, execution consistency, and long-term workflow scalability.

Popular code quality & workflow skills

View category →

grill-me

mattpocock/skills

A relentless interview to sharpen a plan or design.

Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.

handoff

mattpocock/skills

Compact the current conversation into a handoff document for another agent to pick up.

domain-modeling

mattpocock/skills

Build and sharpen a project's domain model. Use when discussing codebase terminology, writing or editing a GLOSSARY.md, or recording or editing an ADR.

codebase-design

mattpocock/skills

Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.

diagnosing-bugs

mattpocock/skills

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.