| name | built-planner |
| description | Comprehensive task planning using the `everything-claude-code` ecosystem. Use this skill when the user has provided detailed specifications or after a specification phase is complete. It spontaneously proposes a detailed execution plan (`task.md`) by consulting the `capabilities_library.md` to select the most appropriate Agents, Skills, and Rules for each step of the task. |
Built Planner
This skill helps you create a detailed execution plan by leveraging the specialized capabilities of the everything-claude-code ecosystem.
When to Use
- Trigger: Spontaneously propose this skill after a specification phase is complete (e.g.,
requirements.md, design.md, or implementation_plan.md are finalized).
- Trigger: When the user explicitly asks for a detailed plan or task breakdown.
- Goal: Transform high-level requirements into a granular, actionable
task.md file.
Execution Steps
1. Analyze Context
Read the completed specifications or requirements provided by the user. Understand the scope, technologies involved, and critical constraints.
2. Consult Capabilities Library
Read references/capabilities_library.md to identify the relevant tools for the job.
- Agents: Which specialized sub-agents (e.g.,
architect, security-reviewer) should be involved?
- Skills: Which workflows (e.g.,
tdd-workflow, backend-patterns) apply to the tasks?
- Rules: Which language-specific rules (e.g.,
typescript, python) must be followed?
3. Generate task.md
Create or update a task.md file with a detailed breakdown. For each main task or sub-task, explicitly state the Recommended Capability (Agent, Skill, or Rule).
Task Format Example
# Task: Implement User Authentication
- [ ] Design Authentication Flow <!-- id: 1 -->
- **Agent**: `architect` (to review the flow against security standards)
- **Skill**: `security-review` (to ensure all checklist items are covered)
- [ ] Implement Backend Logic <!-- id: 2 -->
- **Skill**: `backend-patterns` (for API structure)
- **Rule**: `typescript` (for coding standards)
- [ ] Create User Model <!-- id: 3 -->
- [ ] Implement Login Endpoint <!-- id: 4 -->
- [ ] Verify Implementation <!-- id: 5 -->
- **Agent**: `security-reviewer` (to scan for vulnerabilities)
- **Skill**: `tdd-workflow` (ensure tests are written first)
4. Detailed Instructions
For complex steps, read the specific reference file (e.g., references/skills/tdd-workflow/SKILL.md) to provide granular instructions in the checklist or accompanying comments.