Skip to content
Get packs

Turn Chat into a Spec

Turns the current conversation into a structured spec and publishes it to your project issue tracker.

Source Download skill
Repository
mattpocock/skills
Stars
309,658
Licence
MIT
Files
2 files

What it does

When invoked, the agent stops questioning you and instead summarises what you have already discussed and the codebase it has explored into a formal spec. The spec covers the problem, solution, a long numbered list of user stories, implementation decisions, testing decisions, out-of-scope items and notes. It then publishes the spec to your project's issue tracker and applies a 'ready-for-agent' label. It also asks you to confirm proposed test seams before writing.

Great for

  • Developers who have just finished planning a feature in chat and want a tracked issue written from that discussion
  • Teams using a Matt Pocock-style skills setup who already run the /setup-matt-pocock-skills configuration
  • People who want an agent to hand off well-defined work to another agent via a ready-for-agent label

Skip it if

  • Non-developers or projects without an issue tracker, since publishing depends on tracker configuration that must already exist (the skill tells the agent to ask you to run setup otherwise)
  • Quick brainstorming where you do not want an issue published, because the skill writes to a shared tracker automatically
  • Projects that need the agent to ask clarifying questions first, because this skill explicitly skips interviews and may produce a spec based on incomplete understanding

Try asking

  • “Turn what we just discussed about the checkout redesign into a spec and publish it to the issue tracker”
  • “Write up the spec for the new notifications feature and label it ready for the agent”
  • “Synthesize our planning conversation into a numbered user-story spec and publish it”
Step-by-step guide: how to use Turn Chat into a SpecInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    to-spec.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip to-spec.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip to-spec.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Have the planning discussion first, so the conversation contains the decisions you want captured, since the skill does not ask questions
  • Review the spec and the proposed test seams before or after publishing, because it is written to a shared tracker without an interview step
  • Make sure the issue tracker and triage labels are configured first to avoid the agent stopping to ask you to run setup

What's inside

  • agents/openai.yaml
  • SKILL.md

Redistributed unmodified from mattpocock/skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.