← ClaudeAtlas

task-intakelisted

Mandatory alignment gate before modifying code, files, or configuration — restate the requirement, surface assumptions, enumerate scenarios with expected outcomes, then STOP and wait for explicit user approval before writing any implementation code. Use at the start of any task that involves changing code, files, or configuration in a repository, especially multi-file or multi-step changes. Skip the gate only for pure Q&A, pure explanation, or tiny changes where the user explicitly said "just do it" — the implementation invariants still always apply. For vague greenfield ideas that are not yet concrete tasks, use idea-to-spec first; for triaging issues in unfamiliar codebases, use progressive-abstraction.
HelloWorldU/skill-collection · ★ 0 · AI & Automation · score 72
Install: claude install-skill HelloWorldU/skill-collection
# Task Intake: Alignment Before Implementation ## Goals - Align on requirements, assumptions, scenarios, and expected outcomes *before* acting. - Obtain explicit user approval through an alignment gate. - Ensure implementation starts with clear acceptance criteria. ## Constraints - **Without explicit user approval, write no implementation code.** Stages 1-3 produce text, written in the reply body. - **Ask and align in prose, not through tool calls.** Put questions and confirmations directly in the reply; the user answers next turn. ## Inputs Any request involving modification of code / files / configuration. Pure Q&A, pure explanation, or tiny changes the user explicitly fast-tracked → stages 1-3 may be skipped; **the Stage 4 invariants always apply**. ## Steps ### Pre-check: Branch Sync Before anything else, check whether the current branch is behind the remote main. Work on the latest base — discovering conflicts at push time is too late. ### Stage 1 — Requirement Clarification + Reading List 1. **Restate the requirement** in your own words and list your **assumptions**. Ask immediately about any ambiguity; never decide on the user's behalf. 2. List the **documents and files to read**: the nearest `AGENTS.md`, relevant docs, existing skills, and every source file that will be touched. 3. Read them before advancing to Stage 2. ### Stage 2 — Scenarios & Expected Outcomes Break the task into **concrete scenarios**, each with an **expected outcome**. Must cover: