From great_cto
Classifies a product as SaMD per IMDRF, proposes FDA regulatory path, searches for predicates, and identifies IEC 62304 / ISO 14971 gaps.
How this command is triggered — by the user, by Claude, or both
Slash command
/great_cto:samd-classify [slug] — optional ARCH slugsonnetThis command is limited to the following tools:
The summary Claude sees in its command listing — used to decide when to auto-load this command
You are the great_cto **/samd-classify** command.
## Step 1 — ARCH
## Step 2 — Invoke fda-reviewer
`subagent_type: fda-reviewer` with prompt:
> Classify product in `$ARCH` as SaMD per IMDRF framework.
> Output: proposed Class (I–IV), proposed FDA path (510(k) / De Novo / PMA / exempt), predicate candidate (openFDA 510(k) search), IEC 62304 software safety class proposal, ISO 14971 risk-file scope.
> Write `docs/sec-threats/TM-samd-${SLUG}.md` using `skills/great_cto/templates/TM-samd.md`.
## Step 3 — Surface
Print: proposed class + path, predicate candidate(s), lifecycle gaps, gates...You are the great_cto /samd-classify command.
ARGS="${ARGUMENTS:-}"
SLUG="$ARGS"
[ -z "$SLUG" ] && SLUG=$(ls docs/architecture/ARCH-*.md 2>/dev/null | sort -V | tail -1 | xargs -I{} basename {} .md | sed 's/^ARCH-//')
[ -z "$SLUG" ] && echo "BLOCKED: no ARCH" && exit 1
ARCH="docs/architecture/ARCH-${SLUG}.md"
subagent_type: fda-reviewer with prompt:
Classify product in
$ARCHas SaMD per IMDRF framework. Output: proposed Class (I–IV), proposed FDA path (510(k) / De Novo / PMA / exempt), predicate candidate (openFDA 510(k) search), IEC 62304 software safety class proposal, ISO 14971 risk-file scope. Writedocs/sec-threats/TM-samd-${SLUG}.mdusingskills/great_cto/templates/TM-samd.md.
Print: proposed class + path, predicate candidate(s), lifecycle gaps, gates raised (gate:samd-class, gate:clinical-validation, gate:ide-approval if PMA).
npx claudepluginhub avelikiy/great_cto/validate-ai-deviceAssesses clinical validation for an AI-enabled medical device against QSR (21 CFR 820) and generates a documentation gap report.
/evaluate-ai-riskAssesses an AI feature against EU AI Act high-risk categories, FINRA agent rules, and FDA SaMD classification; produces a regulatory risk screen with launch gates and a phased compliance path.
/audit-ai-systemMaps an AI system against EU AI Act Annex III high-risk criteria and generates a compliance risk assessment with citations, covering risk tier classification, obligations, and remediation.
/invariantsGuides a six-step interview to author and register an architectural invariant catalog entry, then writes validated YAML and shows the effective delta.
/decideRuns an architecture deep research pipeline: scans the repo for patterns, drafts a PRD, live-researches the decision space with citation audits, and produces a handoff-ready decision report.