to-spec

Converts an approved planner.md into a slice-ready spec with testing seams and behaviour boundaries.

2|Updated Jul 24, 2026
One-click install
npx skills add https://github.com/konstantinos-malavazos/claude-code-playbook --skill to-spec-konstantinos-malavazos
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: to-spec
Source: https://github.com/konstantinos-malavazos/claude-code-playbook/tree/main/templates/skills/to-spec
Command: npx skills add https://github.com/konstantinos-malavazos/claude-code-playbook --skill to-spec-konstantinos-malavazos

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When a ticket is too large for a single slice, the approved plan alone lacks the user-facing behaviour boundaries and testing seams needed to cut it into clean slices. This Skill transforms the existing planner output into a compact slice-ready spec without re-interviewing anyone or re-exploring the codebase. ## Core Features & Use Cases - Slice-ready spec generation: Reads planner.md, context-gatherer.md, and ticket-analyzer.md handoffs and writes a slice-spec.md containing the problem, solution, behaviours, and out-of-scope items. - Testing seam identification: Maps each behaviour to the produce/verify step, integration entry point, or /test-ticket step that proves it, flagging shared seams that block clean slicing. - Behaviour boundary definition: Expresses each behaviour as a user story grouped by unit, giving /to-tickets natural lines to slice along. - Use Case: During the /start-ticket DECOMPOSE path, after the planner sets slice-count > 1 and the plan is approved, run this Skill to produce the spec that /to-tickets consumes to split a large ticket into slices. ## Quick Start Run the to-spec skill on the approved planner.md handoff for the current ticket to generate the slice-ready spec before running /to-tickets.

Frequently Asked Questions about to-spec

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

FAQPage Schema
How do I split a large ticket into slices in Claude Code?

Use the /start-ticket DECOMPOSE path: after the planner sets slice-count greater than 1 and you approve the plan, the to-spec skill writes a slice-ready spec, and /to-tickets then cuts the slices from it.

What is a testing seam in ticket decomposition?

A testing seam is the produce/verify step, integration entry point, or /test-ticket step that proves a behaviour works. The spec maps each behaviour to its seam, since behaviours sharing one seam usually cannot be sliced apart cleanly.

When should the to-spec skill not be used?

Do not use it for a normal single-slice ticket, where the pipeline proceeds without decomposition. It only runs in the DECOMPOSE path after the planner sets slice-count greater than 1 and the plan is approved.

Does to-spec re-explore the codebase or interview stakeholders?

No. It is a thin transform of briefs that already exist: it reads only planner.md, context-gatherer.md, and ticket-analyzer.md. If a needed fact is missing, it names the gap and stops rather than re-running exploration.

Can to-spec write the spec to Jira or another tracker?

No. It writes only an internal file at handoffs/<TICKET-ID>/slice-spec.md and never publishes to a tracker or to memory. One ticket ID stays one ticket, and the durable record is written when the ticket lands.