What does this skill do, and when should you use it?
define-goal helps users clarify a concrete objective before work begins, particularly when they want to create a goal, set an outcome, or define success criteria. It structures an objective around the desired result, the relevant artifact or environment, verification evidence, scope boundaries, and when to stop and ask for clarification. The skill favors meaningful metrics and observable proof over generic activity descriptions. It covers goal definition and goal-tool creation only; it does not manage durable snapshots, decision logs, or long-running execution artifacts.
Determines whether goal definition is needed; rewrites an intention as a concrete objective; adds the outcome, relevant artifact or environment, verification method, scope, and stopping condition; introduces tests, thresholds, evidence counts, or other quantitative criteria when appropriate; checks active goal state before creation and creates a goal when the objective meets the quality bar.
- A developer is about to implement a change but needs explicit acceptance tests and scope first.
- A project owner wants to turn “improve quality” into a goal with measurable thresholds and evidence.
- A researcher needs to define the decision, source scope, and evidence standard the research must support.
- An operator needs to specify healthy state, monitoring window, failure threshold, and escalation criteria.
How do you install this skill?
- Before use, verify the goal tool's actual interface, persistence behavior, and recovery behavior on failure.
- Do not treat the skill's rewritten objectives as validated acceptance criteria without user confirmation of scope, metrics, and evidence.
- The repository is marked deprecated; verify the skill's current maintenance source, version, and license.
This is a curated skill. Install it in Codex with $skill-installer define-goal, then restart Codex so the skill is loaded.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/openai/skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/.curated/define-goal" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- improve the checkout flow
Use it when goal definition is needed, for example: Rewrite “improve the checkout flow” as a measurable, verifiable goal with scope and acceptance criteria. If the request is ordinary implementation work without a need for goal definition, perform the work directly.
What are this skill's strengths and limitations?
- Prioritizes verifiable outcomes, explicit scope, and stopping conditions.
- Supports domain-appropriate tests, quality thresholds, performance metrics, and evidence counts.
- Checks for an existing matching active goal before creating another one.
- Covers goal definition and goal creation only; it does not manage snapshots, decision logs, or long-running execution records.
- The source provides no evidence of a test suite, scripts, external services, or specific runtime dependencies.
- If a critical metric or validator is missing and cannot be safely inferred, the user must clarify it.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Goal Definition Assistant this page ✓ OpenAI · Official | 56 · Use with care | ★ 28k | 4mo ago | — |
| MVP Minimalist Advisor | 52 · Use with care | ★ 11k | 6mo ago | — |
| deepPPT | 67 · Recommended | ★ 28 | 2mo ago | MIT |
| vibe-check Product Planning Coach | 62 · Recommended | ★ 619 | 3mo ago | MIT |
| Project Status Artifact ✓ Anthropic · Official | 61 · Recommended | ★ 38k | 11d ago | Apache-2.0 |
How did FollowSkills review this skill?
The skill narrowly limits itself to goal definition and goal-tool creation, requires checking active goals, avoids duplicates, and asks the user when ambiguity materially matters. Points are deducted because sensitive-data handling, data-flow disclosure, rollback, and recovery are not specified; creating a goal may still change durable state.
The workflow is internally coherent and covers triggers, quality gates, ambiguity handling, conflicting active goals, and stop conditions. Because this is a static review with no tests or execution evidence, tool availability and failure-path coverage remain uncertain, so the score is capped below 10.
Audience, intended scenarios, non-fit boundaries, and semantic triggers are relatively clear, including the distinction between ordinary implementation and goal creation. Points are deducted for no explicit Chinese-language support or broader environment-fit evidence; the core function does not appear to depend entirely on overseas network services.
The documentation is readable and progressively organized through overview, workflow, quality criteria, quantification heuristics, clarification questions, and examples. Points are deducted for incomplete versioning, changelog, maintenance ownership, update path, and skill-specific license evidence; the repository README also states that the repository is deprecated.
The skill supplies directly usable guidance for rewriting vague intentions into measurable goals with verification criteria and stop conditions. No representative outputs or execution evidence are included, and comparative benefit over manual goal definition is unverified, so the static ceiling applies.
The skill file, agent configuration, and repository guidance provide auditable primary evidence for the stated scope and behavior. There is no committed test suite, CI coverage, or third-party execution evidence, preventing a higher static verifiability score.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →