swcstudiospace/agent-swarm/a02-requirements
Trust tier 3
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
A02 REQ Requirements Engineer — Turns briefs into a requirements spec, user stories and machine-checkable acceptance criteria (Given/When/Then). Use at the start of any feature or when criteria are ambiguous/untestable. Use when the swarm assigns capability req.elicit, req.spec, req.acceptance.
Install
npx vskill@latest install swcstudiospace/agent-swarm/a02-requirementsSKILL.md
Requirements Engineer (A02 REQ) — omp
When to Use
- The assignment capability is one of: req.elicit, req.spec, req.acceptance.
- The user names this agent, slug
a02-requirements, or codeREQ.
Don't use for: Do not design architecture or write product code.
Procedure
- Dispatch via the omp
tasktool withagent: a02-requirements(definition:omp/agents/a02-requirements.md), passing the signedtask.assignpayload. - The agent finishes by calling
yieldwith atask.resultobject; read it from the task result. - Don't use it for: Do not design architecture or write product code.
Registered tools
Discover with swarm_catalog, call with swarm_tool_call (the caller is your own identity).
req_gwt_check(READ): Count Given/When/Then acceptance criteria in a text block.req_invest_score(READ): Score one user story against the six INVEST checks using its text, dependencies, estimate and criteria.req_ambiguity_scan(READ): Flag modal, quantifier and vague terms in requirement text and score the share of sentences that are ambiguou…req_story_duplicates(READ): Find user stories whose normalized token overlap (Jaccard similarity) meets a duplicate threshold.req_wsjf_rank(READ): Rank backlog items by WSJF from supplied value, criticality, risk and job size, and name the prioritization m…req_criteria_testability(READ): Measure the manual-only share of acceptance criteria and list criteria that can be verified neither automatic…req_traceability_coverage(READ): Compute requirement-to-artifact traceability coverage from requirement and link id lists, reporting uncovered…req_change_impact(READ): Estimate tasks at risk and committed-scope change ratio from the requirement ids a change affects and the req…req_nfr_coverage(READ): Check that a set of NFRs covers the required categories (perf, security, a11y, i18n, compliance) and list the…req_pii_scan(READ): Detect email, phone, SSN, Luhn-valid card and IPv4 patterns in brief or story text and return a redacted copy…req_clarification_window(READ): Account clarification rounds against the 2-round, 48 h rule from supplied timestamps and report overdue round…req_conflict_matrix(READ): Detect contradictory structured requirements on the same subject and attribute and emit the conflict matrix,…
Skills
Read the one that fits with its link; they are hidden from the skill listing.
skill://a02-requirements--acceptance-criteria: Use when a story has no acceptance criteria or its criteria are weak, unstructured or manual-only; write Given/When/Then and prov…skill://a02-requirements--backlog-prioritization: Use when a drafted backlog needs an order; choose MoSCoW or WSJF by size and escalate equal-rank stakeholder ties with an option…skill://a02-requirements--brief-intake: Use when a new project.brief or change.request arrives, before any story is drafted: redact PII, score ambiguity and list open qu…skill://a02-requirements--change-control: Use when requirements change after the spec is PLANNED; estimate impact, raise a requirements.change and require approval above 2…skill://a02-requirements--clarification-round: Use when ambiguity exceeds 0.4 or over 30% of criteria are manual-only; issue a clarification.request within the 2-round, 48 h li…skill://a02-requirements--compliance-tagging: Use when a brief names regulated data or a regime (GDPR, CCPA, HIPAA, PCI-DSS, SOC2); map it to non-negotiable NFRs and prove the…skill://a02-requirements--conflict-resolution: Use when two requirements contradict; publish the conflict matrix, exclude blocked items with rationale and escalate when they bl…skill://a02-requirements--escape-retro: Use when a test escape or incident.alert points at a missing or wrong requirement; find the root cause in the criteria and draft…skill://a02-requirements--nfr-elicitation: Use when a spec has no NFRs or misses a category (perf, security, a11y, i18n); propose measurable targets and reject unmeasurable…skill://a02-requirements--provisional-spec: Use when both rounds are used and one went 48 h unanswered; publish a PROVISIONAL spec with assumption log, per-item confidence a…skill://a02-requirements--publish-gate: Use when the spec is about to be marked VALIDATED; run lint, GWT, traceability, NFR coverage and the DoD check, then publish or b…skill://a02-requirements--story-authoring: Use when a cleaned brief is ready and user stories must be drafted, scored for INVEST and de-duplicated before criteria are writt…
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingomp/skills/a02-requirements/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/swcstudiospace/agent-swarm/a02-requirements)