planner_skill

This skill orchestrates planning and execution for complex tasks, guiding you through structured milestones and actionable workflows.
  • Python

538

GitHub Stars

3

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill solatis/claude-config --skill planner

  • CLAUDE.md1.3 KB
  • README.md3.8 KB
  • SKILL.md3.2 KB

Overview

This skill provides an interactive planner and executor for complex, multi-step tasks. It runs two scripted workflows: a 13-step planning workflow to produce reviewed milestones and a 9-step execution workflow to implement and verify those milestones. Use the skill to generate structured plans, convert intent into code diffs, and manage wave-based implementation with built-in quality gates.

How this skill works

When activated, the skill immediately invokes the appropriate script for planning or execution. The planning script guides the user through context discovery, test strategy, approach options, assumption checks, milestone creation, and multiple review gates that convert intent into validated diffs and documentation. The execution script performs wave-aware implementation, reconciliation with existing code, code and documentation quality reviews, and a final retrospective.

When to use it

  • Tasks with multiple milestones and interdependent steps
  • Work that requires architectural decisions and documented rationale
  • Projects benefiting from staged reflection and quality gates
  • When you need formal conversion of design intent into code diffs
  • Wave-based implementation across multiple developers

Best practices

  • Invoke the planning mode for any nontrivial feature or architectural change
  • Answer assumption checks clearly to reduce rework during review gates
  • Follow the script output exactly—each step injects actionable guidance
  • Use the provided conventions (diff, temporal, structural) when producing code
  • Run the execution workflow only after plan approval to ensure traceability

Example use cases

  • Designing and implementing a new service with multiple dependent components
  • Refactoring a large codebase where decisions must be documented and reviewed
  • Coordinating parallel developer waves to deliver a complex feature set
  • Converting a product requirement into verified code diffs and README documentation
  • Running a repeatable rollout with QA gates and a final retrospective

FAQ

Use intent words: planning for 'plan', 'design', 'architect'; execution for 'execute', 'implement', 'run plan'. The corresponding script is invoked immediately.

Can I skip reviews or gates?

Gates are integral to the workflow; skipping them undermines validation. If you truly must, run the script but expect higher risk and manual checks.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational