agentleFS
seancabalseseancabalse/agentic-playbook/skills/start-task/SKILL.md

start-task

Skill0 starsChanged 15 days agoMITView on GitHub

Start a task: read the context, state the assumptions, ask product questions and write a plan with a check for each step. Use at the start of each task.

17 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use review.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • SentrySkill

    review

    Fixture review skill.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 15 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the request has several readings, list them.

    line 23
  2. Distinctive:

    If there's a simpler approach than the ticket implies, say so.

    line 24
  3. Very specific:

    If the request is vague, large or new (no clear goal, users, scope or acceptance criteria), run refine-requirements first, or delegate to …

    line 25
  4. Distinctive:

    Check for owner-only steps (accounts, secrets, DNS).

    line 27
  5. Distinctive:

    Write a short plan where each step has a check:

    line 31

“Common” is measured: kept by at least 6 in 10 of the 15 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Start a task

Playbook files named here: when the project has its own copy at .playbook/local/<same path>, use that copy instead.

Task: $ARGUMENTS

1. Gather context (read before deciding anything)

  1. The ticket, from the project's tracker (Linear, Jira, GitHub Issues, …), using the access and conventions in AGENTS.md → "Tracker": description, acceptance criteria, comments, linked tickets, and the requirement ids it serves. If there's no ticket, restate the request as acceptance criteria.
  2. The rules: AGENTS.md (already loaded) and the sections relevant to this area. Read the Decision Log entries and ADRs that touch it, and the runbook traps for the tools you'll use.
  3. The code that will change, and its tests. Find the existing pattern for this kind of change and follow it.
  4. Current docs for any API you'll use, for the installed version (${CLAUDE_PLUGIN_ROOT}/principles/01-engineering.md §6). Don't rely on memory for fast-moving frameworks.

2. Think before coding

  • List your assumptions explicitly.
  • If the request has several readings, list them.
  • If there's a simpler approach than the ticket implies, say so.
  • If the request is vague, large or new (no clear goal, users, scope or acceptance criteria), run refine-requirements first, or delegate to the product-owner agent and relay its questions. Then continue from its agreed brief.
  • Identify product decisions the PRD or ticket doesn't settle. Ask them now, in one batch: 2–4 options each, the recommended one first with its reason and trade-offs in customer terms. Don't ask about conventional defaults or facts you can check; choose those and say so.
  • Check for owner-only steps (accounts, secrets, DNS). If one blocks the task, say what it is and plan the work around it.

3. Plan

Write a short plan where each step has a check:

1. <step> → verify: <test / command / observation>
2. …
N. Verify in the real app → verify: <acceptance criteria walked, screenshots>

Include data (migration, policies, tests, generated types), UI states (loading, error, offline, empty), copy in the catalogue, accessibility at the largest text size, docs to update, and whether a new build or deploy step is needed.

4. Start

  • Move the ticket to the tracker's in-progress status (the mapping is in AGENTS.md → "Tracker"), and create a branch named from the ticket off the up-to-date default branch. If the ticket or the brief names a different base branch, or an existing branch to continue, use that instead; bring an existing branch up to date with sync-branch first.
  • Then build it with implement (or delegate to the product-engineer agent), smallest verifiable step first. When a step's check fails, fix it before moving on.

Rules: ${CLAUDE_PLUGIN_ROOT}/principles/00-working-agreement.md, 01-engineering.md, 14-product-and-planning.md.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.