Skip to content

Intent Intake

Turns a support chat log into a reviewed five-section intent.md spec for engineering

Stars
40,512
Licence
MIT
Files
1 file

What it does

This skill is an internal requirements intake for non-engineers at a company called Shipyard. When someone pastes a chat log or describes a product problem, the agent asks numbered batches of clarifying questions, records unknowns as open items instead of guessing, and drafts a five-section intent.md document. People then review and sign that document through the team's tracker, and the accepted document becomes the brief that another skill (/launch) uses to generate an engineering spec. The agent drafts and asks; humans verify, submit, and approve.

Great for

  • Support or operations staff at Shipyard who report product problems but cannot write engineering specs
  • Teams that want a structured, reviewable hand-off from non-engineers to engineers using a tracker
  • Anyone adapting the workflow as a template for their own requirements intake (the content is written for Shipyard's internal process)

Skip it if

  • Public users outside Shipyard: it depends on Shipyard's internal repo layout, tracker records, and the harbor and launch workflows, none of which are included in this file
  • Tasks that need code written directly: the skill explicitly produces decisions, not code
  • Situations needing an automatic approval: the agent is designed never to sign or accept on anyone's behalf

Try asking

  • “Paste this support chat log and help me turn it into an intent.md for the product owner”
  • “Continue the open review for the checkout-timeout intent and list what is still unanswered”
  • “Ask me numbered questions about this bug report until the five sections are complete”
Step-by-step guide: how to use Intent IntakeInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    intent.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

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

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip intent.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

  • Paste the raw chat log as-is; the skill is designed to work from raw feedback and records unknowns rather than inventing facts
  • Answer the numbered questions by number (for example '1 yes; 2 not sure') so unknowns are captured as open questions
  • Treat the drafted intent.md as a draft: confirm the wording yourself before submission, since the agent never signs or submits

What's inside

  • SKILL.md

Redistributed unmodified from Yeachan-Heo/oh-my-claudecode 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.