agentsclimarketplace

Goal clarifier

Skill zaixincheng174-ai/codex-agent-governance-skills/core/skills/goal-clarifier

Codex governance skill pack for AI coding agents: lifecycle gates, repo preflight, diff-scope audit, evidence closeout, and capability-delivery checks.

Install
npx -y skills add zaixincheng174-ai/codex-agent-governance-skills --skill goal-clarifier

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Minimal ambiguity-resolution interview for blocked engineering tasks. Adapted from grill-with-docs; use when Goal, Constraints, or Success Criteria are unclear enough to block safe execution, while preserving lifecycle/governance control.

SKILL.md

2.9 KB, as published. Nobody here has run it

goal-clarifier

Source repo: https://github.com/mattpocock/skills Source path: skills/engineering/grill-with-docs/SKILL.md Source commit: b8be62ffacb0118fa3eaa29a0923c87c8c11985c Local adaptation note: Adapted from an open-ended grilling and documentation-updating skill into a narrow clarification loop. It asks only the minimum questions needed to unblock execution and forbids inline documentation writes without explicit approval.

Purpose

Clarify the smallest set of facts needed to proceed safely:

  • Goal: what becomes true when the task is done.
  • Constraints: what must not change or be touched.
  • Success Criteria: what evidence proves the task is complete.

This skill is subordinate to AGENTS.md, project-lifecycle, repo-preflight, strict audit rules, and explicit user instructions.

Activation

Use only when ambiguity blocks execution or would make assumptions risky.

Before asking the user, inspect available repo evidence if it can answer the question cheaply:

  • AGENTS.md or equivalent local instructions.
  • Existing docs, tests, code, prior artifacts, issue text, or user-provided files.
  • Current worktree state when relevant.

Ask one question at a time. Include the recommended answer when useful, but do not ask for confirmation if repo evidence already resolves the ambiguity.

Clarification Loop

  1. State the blocking ambiguity in one sentence.
  2. Check whether repo evidence can answer it.
  3. If evidence answers it, proceed with the evidence and cite the path or command.
  4. If evidence cannot answer it, ask exactly one question.
  5. Stop after the minimum questions required to clarify Goal, Constraints, and Success Criteria.

Forbidden actions

  • Do not run a broad interview when execution is already safe.
  • Do not update CONTEXT.md unless explicit EXECUTE approval is given.
  • Do not create ADR unless explicit EXECUTE approval is given.
  • Do not edit AGENTS, CLAUDE.md, docs/agents, issue trackers, labels, commits, PRs, or comments.
  • Do not publish, deploy, git push, or run setup.
  • Do not use this skill to replace the Goal Contract or lifecycle gates.

Stop conditions

Stop when:

  • Goal, Constraints, and Success Criteria are clear enough to proceed.
  • The next question would be nice-to-have rather than blocking.
  • The user says proceed, no more questions, audit only, or no planning.
  • Local instructions make the next step read-only or execution-blocked.

Output Pattern

Goal Clarifier
Blocking ambiguity: <one sentence>
Evidence checked: <paths/commands or none>
Question asked: <one question or none>
Resolved working assumption: <if evidence supports it>
Stop condition: <why clarification ended>

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.