ccplan

Transform ambiguous requirements into constraint-bound implementation plans with ECL documents.

1|Updated Jul 14, 2026
One-click install
npx skills add https://github.com/TzJ2006/gadget --skill ccplan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ccplan
Source: https://github.com/TzJ2006/gadget/tree/main/skills/ccplan
Command: npx skills add https://github.com/TzJ2006/gadget --skill ccplan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Diverge-then-converge requirement engineering with prompt calibration, adversarial validation, feasibility probing, and an Evolving Constraint Language (ECL) that persists decisions across sessions and models. Replaces linear planning with a spiral: calibrate → hypothesize → challenge → diverge → converge → probe → confront → review → implement → loop.

Core Features & Use Cases

  • Phase-driven planning for ambiguous requirements across multi-stakeholder environments.
  • Persistent context via ECL documents that survive across sessions and models.
  • Integrated sanity checks: prompt calibration, hypothesis interrogation, divergent exploration, adversarial filtering, and feasibility probing.
  • Guarded implementation flow: Review Gate before implementation; auto-generation of feature guards to prevent regressions.

Quick Start

Run /ccplan with your feature request to begin Phase 0 calibration and let the spiral planning workflow guide you through to implementation.

Frequently Asked Questions about ccplan

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

FAQPage Schema
How do I plan implementation for ambiguous multi-stakeholder requirements?

Planning for ambiguous requirements uses a phase-driven diverge-then-converge workflow. It transforms vague requests into constraint-bound implementation plans through prompt calibration, hypothesis interrogation, and feasibility probing.

What is adversarial validation in requirement engineering?

Adversarial validation in requirement engineering is a mechanism that challenges hypotheses during the planning workflow. It filters divergent ideas by interrogating feasibility before converging into a formal implementation plan.

How do I persist planning decisions across multiple AI sessions?

To persist planning decisions across multiple AI sessions, use an Evolving Constraint Language (ECL) document. ECL stores constraint-bound decisions, allowing implementation plans to survive across different sessions and models.

Can I enforce a review gate before feature implementation?

Yes, you can enforce a review gate before feature implementation. The phase-driven workflow automatically escalates to a formal Review Gate, generating feature guards to prevent regressions before implementation begins.

Does linear planning work for complex architectural choices?

Linear planning is replaced by a spiral workflow for complex architectural choices. This approach cycles through calibration, hypothesis interrogation, and adversarial filtering to handle ambiguity better than linear methods.