creating-issueslisted

Authors well-formed harness issue(s) — a single issue or a dependency-ordered roadmap — from the operator's conversation, from a PRD written by grill (docs/prd/<slug>.md), or from a deepening candidate written by proposing-deepening (docs/architecture/deepening-candidates.md). Applies the small-delivery-unit sizing rule, coaches VERIFIABLE acceptance criteria (the locked_tests derive from them), keeps model assumptions attackable and blocks on open questions, wires harness-deps for order, creates issues harness:ready EXCEPT retrofit issues (local-only, unlabelled, never auto-merged), and runs the DAG lint. Use when the operator wants to create an issue or organize a roadmap.
orobsonn/claude-harness · ★ 1 · AI & Automation · score 67

Install: claude install-skill orobsonn/claude-harness

# Creating-Issues — author the pipeline's highest-leverage input The issue is the **last high-bandwidth human control point**: once it is created, the engine runs plan → build → review → merge end-to-end with no human touching the work again (under auto-merge the PR review is post-merge). A few minutes getting the acceptance criteria, scope, size, and dependencies right decides whether an autonomous run produces gold or burns a whole cycle on garbage you only discover at the PR. This skill makes that authoring excellent. **The STANDARD lives in `rules/creating-issues.md` — this skill is the PROCEDURE that applies it. One source of truth: read that rule, do not restate it here.** Announce at start (pt-br): "Vou montar a(s) issue(s) no padrão do harness." All identifiers/reasoning in English. Every message to the operator is **pt-br, product-language** (what ships, what's verifiable) — never engineering. ## When to use - The operator wants to create an issue, or organize a **roadmap** (several ordered/related issues). - Invoked directly (`/creating-issues`), by `grill` at the end of a PRD interview, or when the entry-gate advisory nudges toward the form. - NOT for triaging an already-existing request into a delivery mode — that is `triaging-requests`. ## Interactive vs headless - **INTERACTIVE (local):** ask the clarifying questions below before creating anything. - **HEADLESS (cloud):** never `AskUserQuestion`; derive from the trigger prompt, proceed, and record an