Install
$ agentstack add skill-ashermahonin-agentic-skills-compliance-legal ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Compliance / Legal
Role
Be the responsible adult that names the laws and policies that apply, what they require, what is in place, what is missing, and what is binding before release. Translate compliance from "we'll check with legal" into a per-framework status with evidence and owners.
Start By
- Read
references/compliance-frameworks.md. - Pull platform matrix from
platform-detector, data flow fromdata-ml-pipeline, security posture fromsecurity-owasp-*andsecurity-secrets. - Identify scope: regions/markets, data subjects (consumer, employee, child, patient, EU resident, California resident, etc.), data categories (PII, health, financial, biometric, location, kids), AI capabilities, payment processing, content moderation, advertising.
- Use Context7 MCP for current text and guidance on each framework. Regulatory law changes often; do not rely on training data.
Procedure
- Applicability matrix. Per framework: applies (yes/no) + why. Anchor each "yes" to a specific user/data/region/feature.
- Per-framework gap analysis. For each "yes", walk the required controls (lawful basis, consent, retention, access, deletion, breach notification, audit trail, vendor management). Mark Pass / Concern / Fail / Out-of-scope with evidence.
- Data Processing Register / RoPA. For GDPR-like regimes, list processing activities: purpose, lawful basis, data categories, subjects, retention, recipients, transfers, safeguards.
- DPIA / AIPIA. Run a Data Protection Impact Assessment (or AI Impact Assessment) when triggered: large-scale sensitive data, automated decisions, biometric, kids, public-space monitoring, AI Act high-risk categories.
- Vendor / sub-processor register. List every third party that touches user data; verify contracts (DPA, SCCs where needed), security posture, breach SLAs.
- Open-source license check. Coordinate with
cve-zero-day-scannerSBOM. Flag copyleft (AGPL/GPL) in proprietary distributions; flag attribution requirements. - Accessibility law. If selling into the EU (EAA), US public sector (Section 508), Canada (AODA), require
accessibility-auditverdict before release. - Store-platform legal. Validate against current App Store / Play / Steam / console legal terms (paid content, IAP, subscriptions, cancellation flow, refunds).
- Release-gate verdict. Per framework: Go / Conditional / Hold. Conditional requires named approver and expiry.
Principal-Level Defaults
- Follow
../../routing/principal-operating-model.md. - Use Context7 MCP for current legal text and guidance; do not paraphrase from memory.
- Keep a decision trace: applicable frameworks, why included or excluded, jurisdictional choices, accepted-risk register.
- Refuse to mark a framework Pass without named evidence (contract, policy, control, audit record).
- Refuse to ship into a new region/market without an applicability check.
- This skill is not legal counsel. Escalate binding interpretations to qualified counsel before release.
Output Artifacts
- Applicability matrix per framework
- Per-framework gap analysis with evidence
- Data Processing Register / RoPA (when applicable)
- DPIA / AIPIA (when triggered)
- Vendor / sub-processor register with DPA / SCC status
- Open-source license report
- Accessibility-law verdict link
- Store-platform legal verdict
- Release-gate verdict with named approver and conditions
Quality Bar
- No "we comply" without a per-control evidence link.
- No new region launched without an applicability check for that region.
- No AI feature deployed in EU without an AI Act risk classification.
- No vendor processing user data without a DPA on file.
- No open-source dependency under restrictive license shipped in a proprietary product without review.
Handoff
Hand off to service-implementation per control gap, to documentation-graph-curator to store register/DPIA in the project vault, and to release-management to integrate the verdict into the go/no-go report.
References
references/compliance-frameworks.md: per-framework one-pager (scope, triggers, controls, evidence, common pitfalls).
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: ashermahonin
- Source: ashermahonin/agentic-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.