templates/skills/coder/rules/requirements/SKILL.md
Requirement quality rules for coding tasks with traceable MUST/SHOULD and verifiable acceptance criteria.
npx skillsauth add ozerohax/assistagents coder-rules-requirementsInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
<when_to_use> <trigger>Defining or validating coding task requirements before implementation</trigger> <trigger>Need to enforce definition of done and scope boundaries</trigger> <trigger>Need requirement-to-evidence traceability in reviews</trigger> </when_to_use>
<normative_language> <rule>MUST indicates absolute requirement</rule> <rule>SHOULD indicates strong recommendation with documented exception</rule> <rule>MAY indicates optional behavior</rule> </normative_language>
<input_requirements> <required>Task objective and in-scope boundaries</required> <required>Acceptance criteria with pass/fail expectations</required> <required>Quality gates (tests/lint/typecheck/security/perf as relevant)</required> <required>Constraints (security, compatibility, rollout, tooling)</required> <optional>Out-of-scope and future work notes</optional> </input_requirements>
<requirement_categories> <category>Scope and intent</category> <category>Functional acceptance criteria</category> <category>Definition of done gates</category> <category>Constraints and guardrails</category> <category>Non-functional requirements</category> <category>Verification evidence</category> </requirement_categories>
<writing_templates> <template>[REQ-ID] Subject MUST action under condition; verification command/result; evidence artifact</template> <template>[REQ-ID] Subject SHOULD behavior; allowed exception; compensating control</template> <template>[AC-ID] Given context, When action, Then expected observable result</template> </writing_templates>
<quality_rules> <rule importance="critical">Every requirement is singular, unambiguous, and testable</rule> <rule importance="critical">Every AC maps to direct evidence in code/tests/behavior</rule> <rule importance="high">Out-of-scope changes are flagged explicitly</rule> <rule importance="high">NFRs are measurable when they affect release decisions</rule> </quality_rules>
<checklist> <item>Requirements use explicit MUST/SHOULD/MAY intent</item> <item>Each REQ has verification method and evidence target</item> <item>AC coverage is complete and traceable</item> <item>DoD gates are declared before implementation starts</item> <item>Exceptions include rationale, owner, and expiration</item> </checklist><do_not> <item importance="critical">Do not accept vague language like "improve" without metric</item> <item importance="high">Do not mix requirement and implementation preference without reason</item> <item importance="high">Do not mark done without traceability matrix</item> </do_not>
<output_requirements> <requirement>Requirement table with REQ-ID, type, verification, evidence</requirement> <requirement>AC coverage matrix and uncovered gaps</requirement> <requirement>List of approved exceptions and compensating controls</requirement> </output_requirements>
<references> <source url="https://www.rfc-editor.org/rfc/rfc2119.txt">RFC 2119</source> <source url="https://www.rfc-editor.org/rfc/rfc8174.txt">RFC 8174</source> <source url="https://scrumguides.org/scrum-guide.html">Scrum Guide (Definition of Done)</source> <source url="https://www.incose.org/docs/default-source/working-groups/requirements-wg/guidetowritingrequirements/incose_rwg_gtwr_v4_summary_sheet.pdf">INCOSE Requirements Writing Guide</source> <source url="https://iso25000.com/index.php/en/iso-25000-standards/iso-25010">ISO/IEC 25010 Overview</source> <source url="https://owasp.org/www-project-application-security-verification-standard/">OWASP ASVS</source> </references>testing
Visual regression (baseline diffs, UI stabilization)
testing
Design detailed reproducible test cases (Given/When/Then, one check per case); not smoke/regression checklists
development
Mobile testing (native/hybrid/web, gestures, offline, interruptions)
testing
Load testing (k6/JMeter scenarios, benchmarks)