
Create Implementation Plan
Writes structured, checkbox-based implementation plans for complex tasks without changing code.
- Repository
- tailcallhq/forgecode
- Stars
- 7,638
- Licence
- Apache-2.0
- Files
- 6 files
What it does
When you ask an AI agent for a plan, roadmap, or strategy for a complex change, this skill makes it research your codebase first and then write a Markdown plan file with objectives, checkbox task lists, verification criteria, risks with mitigations, and alternative approaches. The agent is told not to edit your code, write code snippets in the plan, or run tests or builds. After writing the plan, it runs a bundled validation script to check the file's naming and structure and fixes any problems it reports.
Great for
- Developers who want a written plan reviewed before an AI agent starts coding
- Teams breaking a large feature into ordered, checkable steps
- People who need a risk and alternatives analysis before committing to an approach
Skip it if
- Simple one-line fixes where a plan adds overhead without value
- Situations where you want the agent to implement changes directly, since the skill forbids code modifications
- Cases where you need the plan to contain concrete code, because the skill explicitly prohibits code in plans
Try asking
- “Create a plan for adding OAuth login to this app, with risks and alternatives”
- “Give me a roadmap for migrating our database layer to a new ORM”
- “What is a strategic plan for splitting this monolith service into modules?”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 6 original files, the Apache-2.0 licence and an attribution note.
create-plan.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip create-plan.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip create-plan.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Name the task and scope clearly so the agent can cite the right files with filepath:line references
- Review the generated plan in plans/ before asking for implementation, since the skill suggests switching to an implementation agent afterward
- Check that the validation script passed; the skill requires it and it catches naming and structure errors
What's inside
- README.md
- references/example-plan.md
- references/plan-template.md
- SKILL.md
- validate-all-plans.sh
- validate-plan.sh
Redistributed unmodified from tailcallhq/forgecode under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.