goal-setter
Draft, audit, or activate a compact /goal when the user asks for a persistent objective or wants Codex to work until a verifiable outcome is true. Defines Done, evidence, constraints, stop conditions, optional one-question-at-a-time clarification, and only necessary worker use. Not for ordinary implementation, Q&A, one-off edits, loose brainstorming, or subjective work with no rubric.
npx skills add gotalab/goal-setter-skill --skill goal-setter --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Turn a rough request into a compact /goal that states the expected result, how to verify it (Done), any constraints, stop conditions, and only the necessary worker use. It treats this as Goal intake and decides whether to ask questions, draft, activate, or suggest a normal prompt.
How it works
The skill drafts a Goal that explicitly defines the final state, evidence for verification, and pass/fail criteria. It emphasizes verification targets over detailed procedures, starts from a minimal prompt that preserves the requested outcome, and uses the runtime to emit the exact /goal … line when appropriate. If the request is rough, it may prompt for clarification first following a structured process (answer three plain questions, reconstruct user intent, and decide whether to draft or ask clarifying questions). It includes guidelines for when to use a preliminary Goal if verification is not yet possible, and instructs how to compress the goal around key elements (outcome, evidence, constraints, boundaries, iteration policy, and stop conditions).
When to use it
Use when the task has a single durable objective that may require multiple iterations and its Done state can be verified via commands, artifacts, diffs, benchmarks, or a written rubric. It should avoid scenarios that are subjective or lack a clear rubric. If verification is missing or ambiguous, it can propose a preliminary Goal to establish checks before proceeding.
What it can touch
The skill operates through the runtime’s native mechanism to set the goal and draft the required /goal line. It references the need to emit the exact /goal … line and to avoid claiming the goal was set unless it was.
Caveats
The guidance emphasizes not shrinking the outcome, prioritizing concrete verification and constraints, and avoiding implementation details in the Goal itself. It requires careful compression around six elements and may spawn subagents for verification if needed to satisfy Done.
# Goal Setter Turn a rough request into a compact `/goal` that says what result is expected, what Done means, how to check it, what must not be broken, when to stop, and how Codex should run it. Treat this as Goal intake: decide whether to ask, briefly explore, draft, activate, or say a normal prompt is a better fit. A Goal states the user's requested outcome and how to know it worked; it is not an implementation recipe. Favor verification targets and feedback loops over detailed procedure rules: long rule sets can fit sample tasks while failing real work, but outcome, evidence, and stop constraints travel across projects. Never shrink or reinterpret the outcome; minimize only the surrounding prompt. Start from the smallest prompt that preserves the requested product/task outcome, then add only clauses that change execution, verification, safety, or output. Default to an inline condition. Set the goal through the runtime's native mechanism, or emit the exact `/goal …` line; never claim it was set unless it was. ## When to use A Goal fits when the task has one durable objective, may take many iterations, and Done can be verified by commands, artifacts, diffs, screenshots, benchmarks
- When to use
- Before drafting
- What the goal should contain
- Parallel (decomposable work)
- Activate
- Readiness check (before activating)
What does the goal-setter skill do?
Draft, audit, or activate a compact /goal when the user asks for a persistent objective or wants Codex to work until a verifiable outcome is true. Defines Done, evidence, constraints, stop conditions, optional one-question-at-a-time clarification, and only necessary worker use. Not for ordinary implementation, Q&A, one-off edits, loose brainstorming, or subjective work with no rubric.
How do I install it?
Run `npx skills add gotalab/goal-setter-skill --skill goal-setter --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From gotalab/goal-setter-skill, a repository with 94 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.