MCP App Store
by KeenEthics
SEAT Security Awareness Assessment icon

SEAT Security Awareness Assessment MCP Server for Claude

By Roonik Films, LLC

Overview

SEAT measures a security awareness program across four pillars, Strategy, Engagement, Assessment and Training, each on a 1 to 5 maturity scale. This connector lets Claude administer the 21-question assessment one pillar at a time, score the result, and show the five questions holding the program back along with what the next level looks like. From there it maps the result to any of 16 compliance and AI governance frameworks, including NIST CSF 2.0, ISO 27001, NIS2, DORA, CMMC, PCI DSS 4.0, SOC 2, GDPR, the EU AI Act and NIST AI RMF. Each requirement comes back met or gap, with missing evidence listed for the gaps. Requirements the core questions cannot score are reported as not assessed, never as gaps. It also serves eight published research reports (Verizon DBIR 2026, IBM Cost of a Data Breach 2026, SANS 2026, CrowdStrike 2026 and others) mapped to SEAT outcomes, so a weak score can be read against outside evidence. Research explains exposure. It never changes a score or decides compliance. No account and no sign-in. Answers are numbers only, processed in memory and not stored. Every tool is read-only. Content comes unmodified from the public seat-assessment repository under CC BY-ND 4.0.

How to connect

  1. Claude: Open the SEAT Security Awareness Assessment listing in the Claude connectors directory and choose Connect. Claude listing SEAT Security Awareness Assessment exposes 8 tools here.

Tools

seat_build_program_plan

Claude
Builds the SEAT program plan for a completed assessment, using the same generator as the SEAT app: an executive summary with up to three focus areas, a crawl, walk and run roadmap of initiatives (each with what it includes, the deliverable and the success metric), plan-level recommendations, what is working and the gaps, and a program foundation template. Requires all 21 answers. Optional context sharpens the plan: frameworks in scope, budget direction, how much of a full-time role runs the program, an unpublished policy, fragmented tooling, and global operations. Answers and context are processed in memory and not stored.

seat_compare_assessments

Claude
Shows progress between two complete SEAT assessments, for example a baseline and a re-assessment six months later: overall and pillar movement with level changes, which outcomes moved, every question whose answer changed (with the option text before and after), and, given a framework id, the alignment change with gaps closed and new gaps. Each assessment is passed as the answers_code returned by seat_score_assessment (for example "v1.323223223434312143434") or as the 21 answers. Nothing is stored; the respondent keeps their codes.

seat_get_facilitation_notes

Claude
Returns the facilitation guide for one SEAT question: why it exists, what strong looks like, and the trap that makes people over-score. Used to give a respondent context before a question, or to help when they are between two levels.

seat_get_guide

Claude
Practitioner guides from HumanRisk on running a security awareness program: measuring beyond vendor dashboards, writing the first program document, building stakeholder support, budget conversations, security champions, HRM governance and more. Without a slug, lists the guides (optionally only those for one pillar) with title, summary, audience and link. With a slug, returns that guide in full as markdown. Use after scoring to point the respondent at the guide for their weakest pillar.

seat_get_overview

Claude
Returns what the SEAT assessment is, the four pillars and their question ids, the five maturity levels with score ranges, and the published scoring rules. Use it to start a SEAT assessment or to answer how SEAT is scored. Takes no input.

seat_get_questions

Claude
Returns the SEAT core instrument questions with all five answer options each, verbatim. Pass a pillar to get one pillar at a time (recommended for administering the assessment), or omit it for all 21. Question text is licensed CC BY-ND 4.0 and is returned unmodified.

seat_get_recommendations

Claude
Returns the SEAT recommendations that apply to a completed assessment: for each pillar, weakest first, what the score signals (observation), why it matters (strategic insight), the next concrete step, and a stretch goal. These come from the SEAT app library, the same guidance the hosted assessment shows, matched to the 21 answers with the app's own rules. Use it after seat_score_assessment, when the respondent asks what to do next. Requires all 21 answers. Answers are processed in memory and not stored; the server does not log answer contents.

seat_get_research_source

Claude
Returns one research source file, verbatim: its findings mapped to SEAT outcomes, with the exact figures and bases that may be cited. The text states which figures may be cited, with their bases, publisher and coverage window.

seat_list_frameworks

Claude
Lists the 16 compliance and AI governance frameworks that SEAT results can be mapped to (NIST CSF 2.0, ISO 27001, NIS2, DORA, CMMC, PCI DSS 4.0, SOC 2, GDPR, MITRE ATT&CK, CRI, EU AI Act, NIST AI RMF, ISO 42001, OWASP LLM, OWASP ASI 2026, MITRE ATLAS) with ids, versions and row counts. Use the id with seat_map_to_framework.

seat_list_research_sources

Claude
Lists the published third-party research reports (Verizon DBIR 2026, IBM Cost of a Data Breach 2026, SANS 2026, CrowdStrike 2026, and others) that are mapped to SEAT outcomes, with publisher, evidence type, coverage window and the outcomes each is best for. Also returns the shared rules for using research: it explains exposure and never changes a score or establishes compliance status.

seat_map_to_framework

Claude
Maps a completed SEAT assessment to one framework using the published adapter rules: per-row thresholds (Required 4.5, Expected 2.5, Recommended 1.5), met or gap with no partial state, not-assessed rows excluded from the score, missing evidence for gaps, and a 0 to 5 weighted alignment score. Requires all 21 answers and a framework id from seat_list_frameworks. Answers are processed in memory and not stored; the server does not log answer contents.

seat_score_assessment

Claude
Scores a completed 21-question SEAT assessment. Returns per-pillar scores and maturity levels with the published level descriptions, the overall score and level, the lowest pillar, the spread between pillars, and the five lowest-scoring questions with the next-level option for each. Requires all 21 answers. The numeric answers are processed in memory and are not stored; the server does not log answer contents. Requires every answer to come from the respondent.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/seat-assessment"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/seat-assessment)