red-team-verifier-patrick-munro
Adversarial verification for AI-generated legal content with systematic fact-checking, source validation, and quality control. Use when User requests verification of legal documents, fact-checking of regulatory content, red team review, or quality assurance before distribution to clients/stakeholders. Provides structured verification reports with severity-categorized errors, verified sources, and distribution readiness assessment.
What this skill does
# Legal Red Team Verifier ## Purpose This skill provides systematic adversarial verification of AI-generated legal content to ensure factual accuracy, proper legal citations, and appropriate disclaimers before distribution to clients or stakeholders. It addresses the #1 concern about AI in legal practice: "How do I know this is accurate?" ## When to Use This Skill Use the Legal Red Team Verifier when User requests: - Verification of AI-generated legal content before client/stakeholder distribution - Fact-checking of legal snapshots, briefings, or analyses - Quality control on compliance documents, regulatory summaries, or legal reports - Red team review of legal outputs (e.g., "verify this", "fact-check this", "red team this document") - Adversarial testing of legal claims or arguments **Trigger phrases**: "verify", "fact-check", "red team", "check accuracy", "validate sources", "quality control", "is this correct", "review for errors" ## Core Verification Categories ### 1. FACTUAL ACCURACY - **Regulatory dates and deadlines**: Verify enforcement dates, compliance deadlines, transition periods - **Article/section references**: Confirm regulation articles, statutory sections, directive provisions exist and are cited correctly - **Numerical data**: Validate statistics, percentages, thresholds, financial amounts - **Entity names**: Check correct naming of agencies, authorities, organizations (e.g., BaFin, ESTI, ENISA, European Commission) - **Timeline accuracy**: Verify historical events, legislative milestones, implementation schedules ### 2. LEGAL AUTHORITY CITATIONS - **Primary sources**: Laws, regulations, directives (e.g., AI Act Article 6(2), GDPR Article 25, NIS2 Article 21) - **Secondary sources**: Case law, administrative guidance, regulatory opinions - **Citation format**: Proper legal citation standards (EUR-Lex references, official journal citations) - **Authority hierarchy**: Ensure primary law not confused with guidance or commentary - **Current vs. superseded**: Verify using current version, not outdated provisions ### 3. ARITHMETIC VALIDATION - **Timeline calculations**: Independently calculate compliance deadlines from effective dates - **Percentage calculations**: Verify mathematical accuracy of percentages, ratios, proportions - **Financial calculations**: Check penalty calculations, cost estimates, threshold determinations - **Logical consistency**: Ensure numbers add up across document (e.g., if mentioning "3 categories" verify exactly 3 are listed) ### 4. SOURCE VERIFICATION - **Verifiable claims**: Every factual claim must link to a verifiable source - **Official sources prioritized**: EUR-Lex, official gazettes, government websites, regulatory authority publications - **No unsourced statistics**: Flag any statistical claim without attribution - **No unsourced quotes**: Every quote must have proper attribution - **Cross-referencing**: Critical claims verified against multiple independent sources ### 5. SPECULATION DETECTION - **Opinion vs. fact**: Clearly distinguish editorial opinion from factual legal requirements - **Uncertainty acknowledgment**: Identify areas where legal interpretation is unsettled or debated - **Predictive claims**: Flag statements about future regulatory developments as speculation - **"Likely" and "probably"**: Ensure speculative language is clearly labeled as such - **Editorial framing**: Identify where AI has inserted interpretive framing not present in source material ### 6. DISCLAIMER ADEQUACY - **Legal advice disclaimer**: "This is not legal advice" where appropriate - **Jurisdiction limitations**: Clear statement of applicable jurisdiction (e.g., "Based on German/EU law") - **Date/version disclaimer**: Document version/date of regulations cited - **Professional consultation**: Recommendation to consult qualified legal professionals for specific situations - **Regulatory uncertainty**: Disclosure where regulation is pending, in draft, or interpretation unclear ## Verification Methodology ### STEP 1: Initial Content Review - Read entire document to understand scope and claims - Identify all factual claims, legal citations, numerical data, and authoritative statements - Note any missing sources, vague language, or unsupported assertions ### STEP 2: Source Verification (ALWAYS ONLINE) - **MANDATORY**: Use web_search for EVERY factual claim, legal citation, and statistical assertion - Prioritize official sources: - EUR-Lex (https://eur-lex.europa.eu) for EU legislation - Official government websites (.gov, .gov.uk, .bund.de, .europa.eu) - Regulatory authority sites (BaFin, ESTI, ENISA, BSI) - Official gazettes and legal databases - Cross-reference critical claims across multiple sources - Document source URLs for all verified facts ### STEP 3: Arithmetic Verification - Independently calculate all timelines, deadlines, and dates - Verify all percentages, ratios, and financial figures - Check internal consistency (e.g., if document says "3 types" verify exactly 3 are listed) - Flag any mathematical errors or inconsistencies ### STEP 4: Citation Validation - Verify article/section numbers exist in cited regulations - Check that citations match current versions (not superseded provisions) - Ensure proper legal citation format - Confirm quotes are accurate (not paraphrased but presented as quotes) ### STEP 5: Speculation Identification - Flag any predictive statements about regulatory developments - Identify editorial opinions presented as facts - Note areas of legal uncertainty or debate - Ensure speculative content is clearly labeled ### STEP 6: Disclaimer Review - Check for legal advice disclaimers - Verify jurisdiction is clearly stated - Ensure date/version of regulations is specified - Confirm recommendation for professional consultation where appropriate ## Output Structure Provide verification results in the following structured format: ``` # LEGAL RED TEAM VERIFICATION REPORT ## Document Analyzed [Title/description of content verified] ## Overall Assessment **Quality Score**: [1-5 scale, 5 = distribution-ready] **Distribution Readiness**: [READY / NEEDS REVISION / MAJOR CORRECTIONS REQUIRED] **Critical Issues Found**: [Number] **Verification Completed**: [Date/time] --- ## ✅ VERIFIED FACTS [List all factual claims successfully verified with sources] - Claim: [statement] Source: [official source URL] Status: ✅ VERIFIED --- ## ❌ ERRORS REQUIRING CORRECTION ### CRITICAL (Immediate correction required) - **Error**: [Description of factual error, legal misstatement, or arithmetic mistake] **Location**: [Where in document] **Correction**: [What should it say] **Source**: [Correct source URL] ### HIGH (Correction strongly recommended) - **Issue**: [Missing critical disclaimer, regulatory uncertainty not disclosed] **Impact**: [Why this matters] **Recommendation**: [Suggested addition/revision] ### MODERATE (Should be addressed) - **Issue**: [Unsourced statistics, editorial framing as fact] **Impact**: [Credibility/accuracy concern] **Recommendation**: [How to improve] ### LOW (Minor improvements) - **Issue**: [Minor inconsistencies, stylistic issues] **Recommendation**: [Optional enhancement] --- ## ⚠️ UNSUPPORTED CLAIMS [Claims requiring verification or removal] - **Claim**: [Statement made without source] **Status**: Could not verify through official sources **Action Required**: Either provide source or remove claim --- ## 📋 MISSING DISCLAIMERS [Recommended disclaimer additions] - **Location**: [Where to add] **Type**: [Legal advice / Jurisdiction / Date-version / Professional consultation] **Suggested Language**: [Specific disclaimer text] --- ## 🎯 DETAILED FINDINGS ### Factual Accuracy [Detailed analysis of factual claims] ### Legal Citations [Analysis of legal authority citations] ### Arithmetic Validation [Analysis of numerical accuracy] ### Source Quality [Assessment of sources used] ### Speculation & Opinion [Analysis
Related in Security
mac-ops
IncludedComprehensive macOS workstation operations — diagnose kernel panics, identify failing drives, audit launchd startup items, decode wake reasons, triage TCC permission denials, manage APFS snapshots, recover from no-boot. Use for: Mac is slow, slow bootup, won't boot, kernel panic, kernel_task hot, mds_stores CPU, photoanalysisd, cloudd, login loop, gray screen, sleep wake failure, drive failing, IO errors, APFS snapshots eating space, Time Machine local snapshots, Spotlight indexing, launchd, LaunchAgent, LaunchDaemon, login items, TCC permissions, Full Disk Access, Screen Recording denied, Gatekeeper, quarantine, com.apple.quarantine, app is damaged, helper tool, /Library/PrivilegedHelperTools, pmset, wake reasons, dark wake, sysdiagnose, panic.ips, DiagnosticReports, configuration profile, MDM profile, remote diagnostics over SSH.
a11y-audit
IncludedRun accessibility audits on web projects combining automated scanning (axe-core, Lighthouse) with WCAG 2.1 AA compliance mapping, manual check guidance, and structured reporting. Output is configurable: markdown report only, markdown plus machine-readable JSON, or markdown plus issue tracker integration. Use this skill whenever the user mentions "accessibility audit", "a11y audit", "WCAG audit", "accessibility check", "compliance scan", or asks to check a web project for accessibility issues. Also trigger when the user wants to verify WCAG conformance or map findings to a specific standard (CAN-ASC-6.2, EN 301 549, ADA/AODA).
erpclaw
IncludedAI-native ERP system with self-extending OS. Full accounting, invoicing, inventory, purchasing, tax, billing, HR, payroll, advanced accounting (ASC 606/842, intercompany, consolidation), and financial reporting. 413 actions across 14 domains, 43 expansion modules. Constitutional guardrails, adversarial audit, schema migration. Double-entry GL, immutable audit trail, US GAAP.
assess
IncludedAssesses and rates quality 0-10 across multiple dimensions (correctness, maintainability, security, performance, testability, simplicity) with pros/cons analysis. Compares against project conventions and prior decisions from memory. Produces structured evaluation reports with actionable improvement suggestions. Use when evaluating code, designs, architectures, or comparing alternative approaches.
spring-boot-security-jwt
IncludedProvides JWT authentication and authorization patterns for Spring Boot 3.5.x covering token generation with JJWT, Bearer/cookie authentication, database/OAuth2 integration, and RBAC/permission-based access control using Spring Security 6.x. Use when implementing authentication or authorization in Spring Boot applications.
code-hardcode-audit
IncludedDetect hardcoded values, magic numbers, and leaked secrets. TRIGGERS - hardcode audit, magic numbers, PLR2004, secret scanning.