JPeetz/agent-skills / skills/documentation-content/spec-writer
spec-writer
VerifiedUse this skill when a vague feature request, ticket, PRD fragment, or rough idea needs to be turned into a structured specification, technical plan, and ordered task breakdown for AI coding agents. Generates all three output sections immediately, then flags every implicit decision as an inline [ASSUMPTION: ...] marker with an impact rating and correction guidance — no back-and-forth questions needed before drafting. Primary keywords: spec-driven development SDD, feature specification generation, technical planning, task breakdown, assumption flagging, acceptance criteria Given When Then, requirements engineering, feature request refinement, specification writing agent, plan-first development. For all major agentic platforms.
| File | Kind | Size |
|---|---|---|
| CHANGELOG.md | markdown | 864 B |
| LICENSE | other | 1.0 KB |
| SKILL.md | skill md | 8.7 KB |
| evals/evals.json | config | 4.4 KB |
evals/evals.json
View on GitHub{
"schema_version": "1.0",
"skill": "spec-writer",
"evals": [
{
"id": "spec-writer-001",
"name": "Happy path: basic feature spec generation",
"description": "Given a simple feature request ('Add password reset via email'), the skill must produce a complete spec, plan, and task breakdown in a single output, with all three sections clearly delimited.",
"input": "/spec-writer Add password reset via email",
"criteria": [
"Output contains a 'Spec' section with purpose, functional requirements, and acceptance criteria",
"Output contains a 'Plan' section with architecture decisions, data model changes, or API contracts",
"Output contains a 'Task breakdown' section with ordered, self-contained tasks",
"Output contains at least one [ASSUMPTION: ...] marker",
"Output contains an assumptions summary section"
],
"pass": "All 5 criteria met",
"fail": "Any of the 5 criteria is missing"
},
{
"id": "spec-writer-002",
"name": "Non-functional requirements inclusion",
"description": "Given a feature request with implicit performance needs ('Build an admin dashboard showing daily active users'), the generated spec must include non-functional requirements covering performance, security, or storage.",
"input": "/spec-writer Build an admin dashboard showing daily active users",
"criteria": [
"Spec section includes at least one non-functional requirement (performance, security, storage, latency, concurrency, or accessibility)",
"The non-functional requirement is stated explicitly (not embedded in a functional requirement)",
"Plan section addresses the non-functional constraint with a concrete decision"
],
"pass": "All 3 criteria met",
"fail": "Any of the 3 criteria is missing"
},
{
"id": "spec-writer-003",
"name": "Assumption marker correctness",
"description": "Every [ASSUMPTION] marker must include an impact rating (HIGH, MEDIUM, or LOW) and correction guidance explaining when the assumption should be revisited.",
"input": "/spec-writer Let users invite teammates to their workspace",
"criteria": [
"Every [ASSUMPTION: ...] marker contains an explicit impact rating (HIGH, MEDIUM, or LOW)",
"Every [ASSUMPTION: ...] marker includes correction guidance ('Correct this if: ...')",
"The assumptions summary at the end is sorted or grouped by impact level",
"Held assumptions are at least 2 in number"
],
"pass": "All 4 criteria met",
"fail": "Any of the 4 criteria is missing — particularly if any assumption lacks an impact rating or correction guidance"
},
{
"id": "spec-writer-004",
"name": "Edge-case coverage",
"description": "Given a feature request involving boundaries ('Add rate limiting to the public API'), the generated spec must cover edge cases including empty states, error states, boundary conditions, and concurrent access.",
"input": "/spec-writer Add rate limiting to the public API",
"criteria": [
"Spec includes at least 3 distinct edge cases (empty state, error state, or boundary condition)",
"At least one edge case addresses concurrent access or race conditions",
"Acceptance criteria include at least one Given/When/Then that tests a boundary (e.g. 'Given the rate limit is N requests/minute, When the N+1th request arrives, Then the API returns 429')"
],
"pass": "All 3 criteria met",
"fail": "Any of the 3 criteria is missing"
},
{
"id": "spec-writer-005",
"name": "Spec/plan separation",
"description": "The Spec section must remain technology-agnostic. No framework, library, database, or infrastructure choice should appear in the Spec. Technology decisions belong only in the Plan section.",
"input": "/spec-writer Add password reset via email",
"criteria": [
"Spec section contains no mention of specific frameworks, libraries, databases, or infrastructure tools",
"Plan section contains at least one concrete technology decision (framework, library, database, or service)",
"Acceptance criteria in the Spec section do not reference implementation details"
],
"pass": "All 3 criteria met",
"fail": "Any of the 3 criteria is missing (e.g. a framework name appears in an acceptance criterion, or the Plan contains no concrete decisions)"
}
]
}Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |