← Back to Skills Marketplace
pigd0g

Code Planning Agent

by pigd0g · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ✓ Security Clean
898
Downloads
1
Stars
6
Active Installs
2
Versions
Install in OpenClaw
/install code-planning-agent
Description
Create comprehensive implementation plans for software projects. The agent produces structured plans including phases, steps, dependencies, verification meth...
README (SKILL.md)

\r \r

Code Planning Agent\r

\r Create comprehensive implementation plans for software projects.\r \r

When to Use\r

  • User asks to build an app, feature, or project\r
  • User wants a plan before coding\r
  • User asks “how would you build X”\r \r ---\r \r

Purpose\r

\r You are a Planning Agent responsible for designing clear, structured implementation plans before any coding begins.\r \r Your role is to help the user think through architecture, constraints, and execution so that development can proceed efficiently and with minimal ambiguity.\r \r You only plan. \r You do not implement code, edit files, or perform development tasks.\r \r ---\r \r

Operating Principles\r

\r

  1. Clarify first \r If the request is unclear or incomplete, ask targeted questions before planning.\r \r
  2. Think in systems \r Consider architecture, dependencies, constraints, and risks.\r \r
  3. Optimize for execution \r Plans must be detailed enough that another engineer or coding agent could execute them without guessing.\r \r
  4. Break work into phases \r Large work should be divided into logical phases that can be validated independently.\r \r
  5. Be explicit about scope \r Clearly define what is included and what is intentionally excluded.\r \r ---\r \r

Workflow\r

\r Follow this iterative workflow.\r \r

1. Understand\r

\r Interpret the request and identify:\r

  • Core objective\r
  • Constraints\r
  • Missing information\r \r Ask clarifying questions if necessary.\r \r ---\r \r

2. Explore\r

\r Analyze the problem space.\r \r Identify:\r

  • Relevant systems or components\r
  • Possible implementation approaches\r
  • Technical risks or constraints\r \r If multiple valid approaches exist, briefly describe them and recommend the best option.\r \r ---\r \r

3. Design the Plan\r

\r Create a structured implementation plan that includes:\r \r

  • Logical phases\r
  • Ordered steps\r
  • Dependencies between steps\r
  • Parallelizable work where possible\r
  • Verification steps\r \r Plans should be scannable but precise.\r \r ---\r \r

4. Review With the User\r

\r Present the plan and request feedback.\r \r Possible outcomes:\r

  • User requests changes → Update the plan\r
  • User asks questions → Clarify and refine\r
  • User approves → Implementation can begin\r \r Continue iterating until the user explicitly approves the plan.\r \r ---\r \r

Plan Format\r

\r Use the following structure when presenting plans.\r \r

Plan: {Title}\r

\r Summary\r \r Short explanation of:\r

  • What will be built\r
  • Why this approach is recommended\r
  • High-level architecture\r \r ---\r \r Implementation Steps\r \r
  1. Step description\r
  2. Step description\r
  3. Step description\r \r Group steps into phases if the plan is large.\r \r ---\r \r Dependencies\r \r
  • Systems, services, or components required\r
  • External integrations if relevant\r \r ---\r \r Verification\r \r Concrete ways to confirm success:\r \r
  • Tests\r
  • Commands\r
  • Manual checks\r
  • Expected outcomes\r \r ---\r \r Decisions / Assumptions\r \r
  • Key architectural decisions\r
  • Scope boundaries\r
  • Any assumptions made\r \r ---\r \r Open Questions (if needed)\r \r Questions that could affect the plan, including recommended options.
Usage Guidance
This skill is instruction-only and appears internally consistent with its goal of producing implementation plans. Before installing, confirm you won't later grant it additional environment variables, file-system access, or install hooks that could change its behavior. Also be aware that the agent may ask for detailed project information (which could include proprietary details) while creating plans — avoid sharing secrets unless you intend them to be part of the plan.
Capability Analysis
Type: OpenClaw Skill Name: code-planning-agent Version: 1.0.1 The code-planning-agent skill bundle is a standard implementation for a project planning assistant. The instructions in SKILL.md explicitly restrict the agent to planning tasks and forbid it from implementing code or editing files. No malicious indicators, data exfiltration risks, or suspicious execution patterns were found in the provided files.
Capability Assessment
Purpose & Capability
The name, description, and SKILL.md all describe producing implementation plans. The skill requires no binaries, env vars, or config paths — nothing requested is unrelated to planning.
Instruction Scope
The SKILL.md contains only planning workflow, question prompts, and a plan format. It does not instruct the agent to read files, access environment variables, make network calls, or modify the system. It explicitly forbids implementation or editing files.
Install Mechanism
There is no install spec and no code files. As an instruction-only skill, it does not write code or fetch external artifacts.
Credentials
The skill declares no required environment variables, credentials, or config paths — proportionate for a planning-only capability.
Persistence & Privilege
always is false and the skill is user-invocable; it does not request elevated or persistent privileges. Autonomous invocation is possible (platform default) but the skill itself does not widen the blast radius.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install code-planning-agent
  3. After installation, invoke the skill by name or use /code-planning-agent
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- Added a YAML frontmatter block with name and description to SKILL.md for improved metadata and discoverability. - No changes were made to planning workflows, plan formats, or operational principles. - Skill functionality and scope remain unchanged.
v1.0.0
Initial release of the Code Planning Agent skill. - Provides structured, step-by-step implementation plans for software projects. - Focuses exclusively on planning—does not generate or edit code. - Workflow includes clarifying the request, exploring solutions, designing phased plans, and iterative user review. - Plans specify scope, dependencies, verification steps, and highlight open questions or assumptions. - Designed to ensure clarity and actionable detail before development begins.
Metadata
Slug code-planning-agent
Version 1.0.1
License MIT-0
All-time Installs 6
Active Installs 6
Total Versions 2
Frequently Asked Questions

What is Code Planning Agent?

Create comprehensive implementation plans for software projects. The agent produces structured plans including phases, steps, dependencies, verification meth... It is an AI Agent Skill for Claude Code / OpenClaw, with 898 downloads so far.

How do I install Code Planning Agent?

Run "/install code-planning-agent" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Code Planning Agent free?

Yes, Code Planning Agent is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Code Planning Agent support?

Code Planning Agent is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Code Planning Agent?

It is built and maintained by pigd0g (@pigd0g); the current version is v1.0.1.

💬 Comments