aiad-sdd

Plugin aiad-sdd from marketplace aiad-sdd in everssteeve/sdd-mode. Commands read from the repo's manifest — today.

Skills (9)

• context-budget

  Use when estimating or auditing a Context Engineering Budget. Computes the 5 health metrics (M1-M5), produces a health score 0–5/5 and actionable recommendations. Triggered by /sdd context, /sdd exec budget check, /sdd resume.

• drift-detection

  Use when checking that code changes are synchronised with their SPEC (Drift Lock). Cross-references modified files against active SPECs and emits an OK/DRIFT verdict. Triggered by /sdd drift-check, /sdd validate, /sdd audit, pre-commit hook.

• ears-validator

  Use when linting requirements or acceptance criteria for EARS syntax (Easy Approach to Requirements Syntax). Detects forbidden vague words (should, might, fast, user-friendly, intuitive…), multiple SHALL in a single requirement, missing trigger keywords, missing subjects, non-observable verbs and un

• grill-me

  Use when a human gate must be interactive (« grill me ») rather than a static form — one question at a time, the agent proposes its recommended answer, the human validates or corrects. Triggered by /sdd gate --guided and /sdd research --guided. Preserves Human Authorship.

• human-authorship-check

  Use when validating that an Intent Statement, decision or strategic artefact has explicit human authorship — ensures the POURQUOI is not generated, paraphrased or invented by the agent. Triggered by /sdd intent, /sdd spec, /aiad intention.

• reasons-canvas

  Use when structuring a SPEC via the REASONS Canvas (SPDD — Kevlin Henney). Facilitates the capture of intent justification before the standard AIAD SPEC format. Triggered by /sdd spec when the Intent is complex/ambiguous, or on explicit user request.

• regulatory-veto

  Use when validating code, a SPEC or an Intent against the 4 Tier 1 governance agents (AI-ACT, RGPD, RGAA, RGESN). Emits PASS / WARN / VETO with structured remediation. Triggered by /sdd validate, /sdd security, /sdd exec, /aiad gouvernance.

• sqs-scoring

  Use when evaluating a SPEC's quality (Spec Quality Score). Scores the 5 SQS criteria + Test de l'Étranger and decides whether the Execution Gate opens (SQS ≥ 4/5). Triggered by /sdd gate, /sdd split, /sdd exec prerequisite check.

• traceability

  Use when generating, auditing or repairing the machine-verifiable traceability matrix Intent ↔ SPEC ↔ Code ↔ Tests. Reads code annotations (@intent, @spec, @verified-by, @governance), cross-references with .aiad/intents/ and .aiad/specs/, and emits orphan / unimplemented / untraced gaps. Triggered b

Similar plugins