regulatory-prior-art-survey

Surveys binding regulatory obligations and produces schema-validated scope maps, search records, extracts, and registers.

1|1|Updated May 24, 2026
One-click install
npx skills add https://github.com/bm629/agent-skills --skill regulatory-prior-art-survey-bm629
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: regulatory-prior-art-survey
Source: https://github.com/bm629/agent-skills/tree/main/skills/regulatory-prior-art-survey
Command: npx skills add https://github.com/bm629/agent-skills --skill regulatory-prior-art-survey-bm629

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, jsonschema, and includes scripts (resource) and references (resource) components.

What problem does it solve? Before choosing an architecture, a product team must know which laws and standards actually bind it — and a fabricated citation or an unrecorded search is worse than no survey at all. This Skill structures that regulatory prior-art survey so every obligation traces to the issuing body's own published text and every search is recorded as run. ## Core Features & Use Cases - Four-wave survey workflow: mint a regulatory scope map, execute one search angle across primary-law registers and standards catalogs, deep-read one instrument into its obligations, then build the regulatory register with merged mandates and escalated conflicts. - Schema-validated artifacts: every output (scope map, search grid, extract record, register) is checked by a bundled validation script against JSON schemas, so a recorded zero is distinguishable from a search that never ran. - Anti-fabrication discipline: mines only issuing-body text, never quotes paywalled or unreachable instruments, separates authority from binding force, and records sanitization of fetched pages against prompt injection. - Use Case: Before architecting a health-data product sold in the EU and UK, run the survey to produce a register of applicable instruments (GDPR, AI Act, accessibility law), merged architecture mandates, and conflicts flagged for counsel. ## Quick Start Use regulatory-prior-art-survey to map the regulatory obligations binding my product scope and build the regulatory register.

Frequently Asked Questions about regulatory-prior-art-survey

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I survey regulatory obligations before choosing a product architecture?

Run the four-wave procedure: mint a regulatory scope map from the handed classification, execute one search angle against primary-law registers, deep-read each instrument into an extract record, then synthesize the regulatory register. Each artifact is validated with the bundled script before the next wave.

What artifacts does a regulatory prior-art survey produce?

It produces regulatory-scope-map.yaml, one search/<angle_id>.yaml per angle, one extracts/extract-<stem>.md per instrument, and regulatory-register.yaml with a report.md beside it. All are schema-validated and record every query verbatim as run.

How does the survey handle paywalled standards like ISO 27001?

A paywalled or unreachable instrument is recorded with its number, text_retrievable set to paywalled or unreachable, and no quoted text. Naming the instrument and stating that its text costs money to read is treated as a legitimate finding, never paraphrased.

Does the survey distinguish binding law from guidance and contractual standards?

Yes. Every candidate carries both authority (closeness to the issuing body) and binding_force (binding, guidance, voluntary, or contractual) as separate fields. PCI DSS, for example, is recorded as an incorporated standard with contractual binding force.

What happens when two regulations impose conflicting obligations?

Obligations merge only within one ordered dimension, resolving to the strictest stated standard. Disagreements on non-comparable dimensions like consent basis or residency, or constraints that do not intersect, are escalated as conflicts with requires_counsel set true, never silently resolved.

Why does the survey record searches that returned zero results?

A recorded zero is evidence that the search ran, while an omitted cell is indistinguishable from a search that never happened. The 2-D coverage grid records every owed group-source pair with its status, so absence claims in the register rest on receipts.