tlc-spec-driven

Plan software projects into testable specifications and atomic tasks.

1|Updated Feb 21, 2026
One-click install
npx skills add https://github.com/marcius-llmus/skills --skill tlc-spec-driven-marcius-llmus
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tlc-spec-driven
Source: https://github.com/marcius-llmus/skills/tree/main/%28development%29/tlc-spec-driven
Command: npx skills add https://github.com/marcius-llmus/skills --skill tlc-spec-driven-marcius-llmus

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Projects and features often suffer from vague requirements, scope creep, and disconnected implementation steps; this skill provides a structured, traceable workflow that turns goals into testable specifications and atomic tasks, reducing ambiguity and rework.

Core Features & Use Cases

  • Four-phase workflow: Specify, Design, Tasks, Implement+Validate to move from idea to verifiable implementation.
  • Persistent project memory: Creates and maintains .specs artifacts (PROJECT.md, ROADMAP.md, STATE.md) for session continuity, decisions, and blockers.
  • Brownfield and greenfield support: Map existing codebases into STACK/ARCHITECTURE/CONVENTIONS/STRUCTURE/TESTING/INTEGRATIONS artifacts, then plan and execute feature work with clear dependencies and acceptance criteria.
  • Context and model guidance: Manages context loading limits, suggests appropriate model classes for tasks, and enforces WHEN/THEN acceptance criteria for verification.

Quick Start

Say Initialize project to create PROJECT.md and set up the .specs directory so the agent can begin spec-driven planning.

Frequently Asked Questions about tlc-spec-driven

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

FAQPage Schema
How do I break down a software project into testable specifications and atomic tasks?

To break down a software project into testable specifications, use a spec-driven workflow that moves through Specify, Design, Tasks, and Implement+Validate phases, generating artifacts like spec.md and tasks.md to reduce ambiguity and rework.

What is the best way to map an existing brownfield codebase before planning new features?

Mapping an existing brownfield codebase involves generating persistent .specs artifacts that document STACK, ARCHITECTURE, CONVENTIONS, STRUCTURE, TESTING, and INTEGRATIONS to provide traceable context for new feature planning.

How do I maintain session continuity for project planning across multiple agent interactions?

Maintain session continuity by creating and maintaining persistent project memory artifacts like PROJECT.md, ROADMAP.md, and STATE.md within a .specs directory to track decisions, state, and blockers across interactions.

How do you enforce acceptance criteria for feature implementation and validation?

To enforce acceptance criteria for feature implementation, define strict WHEN/THEN conditions within your specification artifacts, ensuring that implementation and validation phases verify against these testable requirements.

Can spec-driven planning manage context loading limits for large codebases?

Spec-driven planning manages context loading limits for large codebases by respecting context boundaries and suggesting appropriate model classes for specific tasks, ensuring the agent operates efficiently within its processing constraints.