sp-planning

Generate repo-scoped task tree plans with exact files, commands, and commit messages.

23|2|Updated Mar 1, 2026
One-click install
npx skills add https://github.com/tta-lab/ttal-cli --skill sp-planning-tta-lab
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sp-planning
Source: https://github.com/tta-lab/ttal-cli/tree/main/skills/sp-planning
Command: npx skills add https://github.com/tta-lab/ttal-cli --skill sp-planning-tta-lab

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

sp-planning prevents rushed or incorrect coding by forcing a structured plan that starts with reading the existing codebase, confirms the exact target project/repo, and produces an implementation-ready task tree with clear files, commands, and commit messages.

Core Features & Use Cases

  • Reality-first exploration: Read relevant files, identify constraints and existing patterns, and check prior art before designing anything.
  • Designer guardrails: No code writing, explicit approval before execution, and design at the structure level (refactor before adding behavior).
  • Strict scope control: Enforces one task → one plan → one confirmed project/repo, and instructs splitting tasks when multiple repos are involved.
  • Worker-ready plan quality: Requires each subtask to include exact file paths, what to do, build/test commands, and a copy-paste commit message.
  • Checkpoint alignment and orientation: Includes a conversational checkpoint to confirm approach before proceeding, and optionally generates an orientation flicknote for complex work.

Quick Start

Use the sp-planning skill to produce a repo-scoped task tree plan by first running ttal project list, then exploring the codebase, aligning on approach, and writing a bite-sized implementation plan with exact files, commands, and commit messages.

Frequently Asked Questions about sp-planning

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create an implementation plan for a multi-repo agent workflow?

To create an implementation plan for a multi-repo agent workflow, you must first confirm the exact target repo, explore existing codebase constraints, and then explicitly split tasks into dependent per-repo execution sequences.

Why do my automated coding agents write incorrect code before understanding the existing architecture?

Automated coding agents write incorrect code when they skip reality-first exploration; enforcing a structured planning phase requires workers to read existing files, identify patterns, and pass conversational alignment checkpoints before designing.

What is the best way to decompose a software task into a repo-scoped execution sequence?

The best way to decompose a software task into a repo-scoped execution sequence is to enforce strict scope gating, translate the task into a single-plan structure, and generate a task tree with exact file paths, commands, and exit criteria.

How do I enforce TDD validation and scope control before an agent starts coding?

You enforce TDD validation and scope control by requiring explicit approval before execution, applying designer guardrails that prevent code writing, and validating the plan against project constraints retrieved via project list commands.

What should a worker-ready implementation plan include for CI integration?

A worker-ready implementation plan for CI integration must include exact file paths, build and test commands, specific commit messages, and clear exit criteria for each subtask to ensure safe, automated execution.

Can I use a single planning task to modify multiple repositories simultaneously?

You cannot use a single planning task to modify multiple repositories simultaneously; strict scope control enforces one task to one plan to one confirmed repo, instructing you to split tasks when multiple repos are involved.