safety-compliance-scoping
Skill1 starsChanged 52 days agoMITView on GitHub
Scope which safety and regulatory compliance frameworks plausibly apply to a system — functional safety standards, medical device classification pathways, industrial safety directives — based on its actual use case and failure consequence, before deep technical or product investment. Use early, whenever a system's use case touches health, safety, or regulated industrial contexts, to identify what compliance burden is coming rather than discovering it after the architecture is locked in. This produces a scoping map, not a certification — a qualified regulatory/compliance professional and, where relevant, legal counsel must own the actual compliance path.
In this repository32 other files it gives its agents.
- agent-system-designSkill
- api-contract-designSkill
- architecture-designSkill
- architecture-optimizationSkill
- ci-speedupSkill
- code-handoffSkill
- cross-ai-handoffSkill
- crypto-storage-auditSkill
- database-designSkill
- dataset-sourcingSkill
- distributed-debuggerSkill
- embedded-constrained-codingSkill
- freedom-to-operate-analysisSkill
- frontend-architectureSkill
- github-masterSkill
- idea-finderSkill
- ieee-paper-drafterSkill
- memory-leak-hunterSkill
- ml-model-selectionSkill
- ml-security-auditSkill
- mobile-platform-architectureSkill
- observability-designSkill
- patent-drafterSkill
- patent-novelty-evaluatorSkill
- performance-optimizationSkill
- project-scaffoldingSkill
- researcher-evaluatorSkill
- researcherSkill
- scale-refactorSkill
- secure-code-reviewSkill
- slow-query-optimizerSkill
- test-architectSkill
The file
Files
Safety Compliance Scoping
The expensive mistake here isn't picking the wrong standard — it's making architecture decisions before anyone checked which standards apply, then discovering the required risk classification or traceability regime demands changes that are painful to retrofit. This skill exists to surface that early, cheaply, before the fact.
Step 1 — Classify the actual use case and consequence of failure first
Everything downstream depends on getting this right: is the system providing decision support to a human who remains in control, or acting autonomously with a direct real-world effect? Is the context consumer, industrial, or medical? What's the realistic worst-case consequence if it fails or gives a wrong answer? This classification — not the technology used to build the system — determines which regulatory framework applies at all.
Step 2 — Identify candidate frameworks by domain, not exhaustively
Name the frameworks plausibly relevant to the Step 1 classification, for example:
- Functional safety: IEC 61508 as the base standard, with domain-specific derivatives (ISO 26262 for automotive, IEC 62304 for medical device software, IEC 61511 for process industry) where applicable.
- Medical device regulatory: a jurisdiction's device classification pathway (e.g., FDA risk-based device class in the US, EU MDR risk class in Europe) — classification usually depends on intended use and risk, not just the underlying technology.
- Industrial safety: relevant machinery or workplace-safety directives for the jurisdiction and industry in question. Name only what's plausibly relevant to the actual use case from Step 1 — a system with no medical or autonomous-actuation component doesn't need the full medical-device framework listed just for completeness.
Step 3 — State each framework's high-level requirement burden
For each candidate framework, summarize at a scoping level what it actually demands: documented risk assessment, requirement-to-test traceability, a specific validation/testing regime, independent audit or certification body involvement. This is enough to scope effort and timeline — not enough to execute compliance from.
Step 4 — Flag the earliest expensive-to-retrofit decision points
Identify where in the system's design certain architecture or documentation decisions become expensive to redo once made — a risk classification arrived at late often invalidates architecture choices made before it, and a system built without traceability from requirements to tests from the start is far more expensive to retrofit than to build in from day one. This early flag is the actual value of scoping before deep investment, not after.
Step 5 — State the limits of this exercise plainly
This produces a scoping map — a reasoned starting point for the actual compliance path, not a certification and not a substitute for one. The real path requires a qualified regulatory/compliance professional, and for anything safety-critical or medical, legal counsel, engaged before any compliance claim is made or any certification process is started.
Output
# Compliance scoping: [system]
## Use-case classification and failure consequence
## Candidate frameworks and why each might apply
## High-level requirement burden per framework
## Earliest architecture/documentation decisions this affects
## Explicit limits of this scoping (not a certification)Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

