Skip to main content
Glama

mcp-kinetic-gain

One MCP server, all twelve Kinetic Gain Protocol Suite specs + the v0.1.0 implementation tooling + the DefenseTech 6-pack semantics (8 tools). Drop into Claude Desktop, Cursor, or any MCP-compatible client with a single config entry. The agent gains 75 tools (47 spec + 16 implementation-preview + 8 DefenseTech + 4 AI Claims Decision Card, v0.9.0): AEO Protocol, Prompt Provenance, Agent Cards, AI Evidence Format, MCP Tool Cards, AI Tutor Cards, Student AI Disclosure, Classroom AI AUP, Clinical AI Disclosure, AI Incident Card, AI Procurement Decision Card, AI Claims Decision Card - plus hash attestation (ed25519), audit-stream event composition + chain verification (offline AND live against a running audit-stream-py via AUDIT_STREAM_URL), cross-spec drift detection, Decision Intelligence preview, and the DefenseTech vault resolver + invariant checkers. New in v0.9.0: the AI Claims Decision Card (InsurTech, claims_card_version) - claims_card_validate, claims_card_inspect, claims_card_sign, claims_card_chain.

This is the agent-facing companion to kinetic-gain-visualizer: the visualizer renders the specs for humans, while this server exposes validation and workflow tools for agents. The optional audit_event_emit tool sends an event to a configured audit-stream service.

Tools

  • aeo_fetch - Fetch the full AEO Protocol declaration at an origin's

  • aeo_inspect - Return a structured summary of an AEO declaration: entity

  • aeo_get_claim - Extract a single AEO claim by ID

  • aeo_well_known_url - Compute the canonical AEO well-known URL for an origin

  • prompt_provenance_validate - Validate a Prompt Provenance JSON document against the v0.1

  • prompt_provenance_inspect - Structured summary of a Prompt Provenance document: prompt

  • prompt_provenance_eval_result - Extract a single evaluation suite's result from a Prompt

  • agent_card_well_known_url - Compute the canonical Agent Card well-known URL for a given

  • agent_card_inspect - Structured summary of an Agent Card document

  • agent_card_tool_disclosure - Return the list of tools an agent declares, with side-effect

  • agent_card_validate - Validate an Agent Card JSON document against the v0.1 schema.

  • ai_evidence_validate - Validate an AI Evidence object against the v0.1 schema.

  • ai_evidence_inspect - Structured summary of an AI Evidence object: claim text

  • ai_evidence_verify_hash - Compute SHA-256 over the canonical UTF-8 form of

  • tool_card_well_known_url - Compute the canonical MCP Tool Card well-known URL

  • tool_card_inspect - Structured summary of an MCP Tool Card: tool identity, safety

  • tool_card_tested_with - Return the tested-LLM entries for a tool, optionally filtered

  • tool_card_validate - Validate an MCP Tool Card JSON document against the v0.1

  • tutor_card_well_known_url - Compute the canonical AI Tutor Card well-known URL

  • tutor_card_fetch - Fetch a Tutor Card from a URL

  • tutor_card_validate - Validate an AI Tutor Card JSON document against the v0.1

  • tutor_card_inspect - Structured summary of a Tutor Card: tutor identity, audience

  • tutor_card_subject_check - Classify a topic against the tutor's subject scope

  • tutor_card_coppa_check - Enforce the spec's COPPA conditional rule: if

  • disclosure_validate - Validate a Student AI Disclosure JSON document against the

  • disclosure_inspect - Structured summary of a Student AI Disclosure: assignment

  • disclosure_verify_artifact_hash - Recompute SHA-256 over a candidate artifact and compare to

  • disclosure_verify_prompt_hash - Verify a single prompt hash in a hashed-mode disclosure

  • disclosure_aup_check - Surface the disclosure's declared policy posture and any

  • aup_well_known_url - Compute the canonical Classroom AI AUP well-known URL

  • aup_fetch - Fetch a Classroom AI AUP from a URL

  • aup_validate - Validate a Classroom AI AUP JSON document against the v0.1

  • aup_inspect - Structured summary of a Classroom AI AUP: policy identity

  • aup_check_compliance - Compare an AUP with a Student AI Disclosure on eight declared

  • clinical_ai_well_known_url - Compute the canonical Clinical AI Card well-known URL

  • clinical_ai_fetch - Fetch a Clinical AI Card from a URL

  • clinical_ai_validate - Validate a Clinical AI Card JSON document against the v0.1

  • clinical_ai_inspect - Structured summary of a Clinical AI Card: system identity

  • incident_well_known_url - Compute the canonical AI Incident Card well-known URL

  • incident_fetch - Fetch an AI Incident Card from a URL

  • incident_validate - Validate an AI Incident Card JSON document against the v0.1

  • incident_inspect - Structured summary of an AI Incident Card: incident identity

  • incident_index_fetch - HEADLINE TOOL, fetch a vendor's

  • decision_card_well_known_url - Compute the canonical AI Procurement Decision Card well-known

  • decision_card_fetch - Fetch an AI Procurement Decision Card from a URL

  • decision_card_validate - Validate an AI Procurement Decision Card JSON document

  • decision_card_inspect - Structured summary of an AI Procurement Decision Card: buyer

  • decision_card_infer_status - Validate each rubric row, then infer a preview decision.status

  • decision_card_to_policy_bundle - Read-only deny preview for rejected, withdrawn, expired, or

  • decision_card_signature_check - Structural check on a Decision Card's signatures[] block

  • incident_affected_walk - Walk an Incident Card's affected block and return every

  • incident_remediation_plan - Map each affected URI in an Incident Card to a recommended

  • attestation_canonical_hash - Compute the SHA-256 canonical-JSON hash of an arbitrary value

  • attestation_verify - Verify an ed25519 Attestation envelope against a body and

  • attestation_inspect - Pretty-print an Attestation envelope with structural

  • audit_event_compose - Build a ready-to-POST audit-stream-py GovernanceEvent

  • audit_chain_verify - Walk an array of GovernanceEvents top-to-bottom and verify

  • audit_event_inspect - Pretty-print one GovernanceEvent with structural validation

  • audit_event_emit - POSTs a governance event to the audit-stream-py instance

  • audit_events_query - GET recent governance events from a running audit-stream-py

  • audit_chain_verify_live - Ask the configured audit-stream service to check its full

  • suite_doc_detect_spec - Detect which Kinetic Gain Suite spec a JSON document is by

  • suite_doc_drift - Structural diff between two versions of the same Suite

  • defensetech_vault_resolve_3axis - Deterministic preview of a DefenseTech 3-axis vault contract

  • defensetech_audit_event_check_invariants - Run three deterministic DefenseTech event checks: CUI

  • defensetech_check_dfars_72h_clock - Calculate the DFARS 252.204-7012 discovery-to-filing 72-hour

  • defensetech_check_cui_distribution_statement - Check that a CUI-Specified+ tier event carries the required

  • defensetech_check_itar_us_person - Check that an ITAR resource event has US-PERSON-VERIFIED (or

  • defensetech_incident_classify_event_type - Given a freeform description of a defense-AI incident

  • defensetech_summarize_cmmc_evidence_bundle - Summarize a CMMC L2/L3 readiness evidence bundle: target

  • defensetech_vault_contract_cross_binding_check - Verify the cross_binding_refs block on a DefenseTech vault

  • claims_card_validate - Validate an AI Claims Decision Card against the bundled v0.1

  • claims_card_inspect - Structured summary of an AI Claims Decision Card: claim type

  • claims_card_sign - Compute the canonical SHA-256 hash of an AI Claims Decision

  • claims_card_chain - Link a new AI Claims Decision Card to its predecessor: sets

Specs with a well-known URL convention (AEO, Agent Cards, Tool Cards) get fetch tools. Specs without one (Prompt Provenance, AI Evidence - these usually travel inline with answers or in repos, not at fixed paths) get parse tools that take a document_json string.

Related MCP server: mneme-ai

Network and data boundaries

URL fetch tools make outbound HTTP(S) requests to user-supplied origins. The server rejects local and nonpublic destinations, rechecks redirects, and limits JSON responses to 1 MB. The optional live audit tools require both AUDIT_STREAM_URL and AUDIT_STREAM_TOKEN in the MCP server's environment. The token must match the audit-stream service's bearer token and have at least 32 visible ASCII characters. Use HTTPS for a remote service; HTTP is accepted only for loopback development. Missing or invalid configuration returns an error without sending a request. audit_event_emit sends kind, caller-asserted source, and payload to the configured service and validates its event receipt. The receipt does not prove durable storage or event completeness. Its shared token authenticates this MCP process, not the source label or tenant. Approve the write before calling it; omit secrets and personal data unless the service is approved for them. Hash-chain verification checks continuity only; it does not prove audit completeness.

Schema validation checks the document shape and selected rules. A passing result is not proof of FERPA, COPPA, HIPAA, FDA, EU AI Act, CMMC, export-control status, or other legal compliance. attestation_verify checks a signature against the public key supplied by the caller; it does not establish the key owner's identity. aup_check_compliance evaluates eight disclosed fields only; it does not check policy scope, tool/vendor requirements, parent consent, or Tutor Cards. A true allowed value is not an authorization grant. DefenseTech tools calculate from supplied data; they do not classify materials or verify a person's status or license. Positive Decision Cards require the live policy engine's identity, scope, action, expiry, and trusted-attestation checks before an enforceable bundle can be created.

The DefenseTech 72-hour clock uses the time an incident was discovered, not when it occurred or when an audit event was created. Supply discovered_at and filed_at with time zones; see DFARS 252.204-7012. A report's actual submission and contract applicability require external evidence.

Start with a task

  • Review a card: pass a parsed document to claims_card_validate, then use claims_card_inspect to summarize its declared decision. Investigate any validation failure before relying on the summary.

  • Compare drafts: call suite_doc_drift with two versions of the same Suite document to see structural changes before publication.

  • Check a signature: call attestation_verify with a public key obtained through a trusted channel; separately confirm that the key belongs to the claimed signer.

  • Record an event: prepare the payload with audit_event_compose, review it for sensitive data, then call audit_event_emit only when the configured audit-stream service and write are approved.

Install

The commands below install the version currently published on npm. To run this source checkout instead, use npm ci, npm run build, then node dist/server.js. Check npm view mcp-kinetic-gain version before relying on a particular source change.

npm install -g mcp-kinetic-gain

Or run without installing via npx:

npx mcp-kinetic-gain

For a controlled deployment, pin a reviewed package version in the client config (mcp-kinetic-gain@<version>) and upgrade deliberately.

Claude Desktop config

Add to your claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/, Windows: %APPDATA%\Claude\):

{
  "mcpServers": {
    "kinetic-gain": {
      "command": "npx",
      "args": ["-y", "mcp-kinetic-gain"]
    }
  }
}

Restart Claude. All 75 tools appear in the tools panel. Try:

"Use aeo_inspect on https://mizcausevic-dev.github.io to summarize the entity declaration, then use ai_evidence_verify_hash to check the content_hash of an evidence object against my candidate text."

CLI mode (v0.5.1+)

The same binary doubles as a Suite JSON validator outside any MCP host. Useful in CI, pre-commit hooks, or local sanity-checks.

# Validate a single document
npx mcp-kinetic-gain validate path/to/ai-entity.json

# Validate a tree of well-known files
npx mcp-kinetic-gain validate ".well-known/**/*.json"

# Multiple paths or globs
npx mcp-kinetic-gain validate cards/clinical-*.json cards/incident-*.json

# Other commands
npx mcp-kinetic-gain --version
npx mcp-kinetic-gain --help

The CLI auto-detects which Suite spec each file belongs to via its top-level version field (aeo_version, clinical_ai_card_version, aup_version, etc.) and validates it against the same zod schemas the MCP tools use. Output is GitHub-Actions-aware: when GITHUB_ACTIONS=true, failures emit ::error:: workflow commands so they surface as PR annotations.

Exit codes:

Code

Meaning

0

Every matched file passed validation

1

At least one file failed validation, failed to parse, or no files matched

2

No file in the input matched a known Suite spec

3

Usage error (missing arg, unknown command or flag)

Running mcp-kinetic-gain with no arguments still launches the stdio MCP server - existing Claude Desktop / Cursor configs are unaffected.

Why one server instead of five?

  • One Claude Desktop config entry instead of five

  • Cross-spec workflows are atomic - an agent can agent_card_tool_disclosure to find a Tool Card URI, then call tool_card_inspect on that URI in the same conversation, all through one server

  • Shared schemas + utilities keep the implementation cohesive

  • Deprecation path - if mcp-aeo-server (the AEO-only predecessor) gets retired, the AEO tools live on here with the same names and contracts

Architecture

src/
├── server.ts              # MCP entrypoint and handler dispatch
├── cli.ts                 # JSON validation CLI
├── tools.ts               # 75 tool descriptors (JSON Schema inputs)
├── schemas.ts             # Zod schemas for the 12 specs
├── common.ts              # guarded fetch and canonical hashes
└── handlers/              # one module per spec plus cross-spec and live tools

Each handler module is independent and could be split into a separate package if needed.

Hash canonicalization

ai_evidence_verify_hash follows the AI Evidence Format spec's canonical SHA-256 rules:

  1. Read content as UTF-8

  2. Normalize line endings to \n

  3. Strip a single trailing newline

  4. SHA-256, lowercase hex, prefixed sha256:

If your candidate_text produces an unexpected mismatch, check CRLF vs LF and trailing newlines first.

Tests

Run the unit and local HTTP tests without an external service:

npm install
npm run typecheck
npm test
npm run build

License

This server: AGPL-3.0. If a modified version serves users over a network, the AGPL requires an offer of Corresponding Source to those users under its terms. See the GNU AGPL FAQ and LICENSE.

The specs themselves: MIT. Maximally permissive. Anyone may implement, validate against, or extend any Kinetic Gain Protocol Suite specification. The dual-license split is deliberate: the protocol stays open, the reference server is copyleft.

Kinetic Gain Protocol Suite

75 tools total across the twelve specs below plus cross-cutting ops (hash attestation, audit-stream events, cross-spec drift) and the DefenseTech tooling. See the Tools catalog above for the full per-tool list (47 spec + 16 implementation-preview + 8 DefenseTech + 4 AI Claims Decision Card).

Spec

Vertical

AEO Protocol

Core

Prompt Provenance

Core

Agent Cards

Core

AI Evidence Format

Core

MCP Tool Cards

Core

AI Tutor Cards

EdTech

Student AI Disclosure

EdTech (FERPA/COPPA)

Classroom AI AUP

EdTech

Clinical AI Disclosure

HealthTech (FDA SaMD + HIPAA)

AI Incident Card

Cross-cutting (EU AI Act Article 73)

AI Procurement Decision Card

Cross-cutting (buyer-side, OMB M-24-10 / NIST AI RMF rubric-friendly)

AI Claims Decision Card

InsurTech

Suite hub: suite.kineticgain.com Companion visualizer: kinetic-gain-visualizer Red-team bench: prompt-injection-bench Reference library: kgp-verify (npm kinetic-gain-protocol, zero-dep MIT hash/sign/verify for KGP)


Connect: LinkedIn · Kinetic Gain · Medium · Skills

Available Tools

75 tools
aeo_fetchA

Fetch the full AEO Protocol declaration at an origin's /.well-known/aeo.json. Returns the raw conforming JSON document.

Inputs: origin (required).

aeo_get_claimC

Extract a single AEO claim by ID. Returns the claim object or a not-found error listing available claim IDs.

Inputs: origin (required), claim_id (required).

aeo_inspectA

Return a structured summary of an AEO declaration: entity, source/verification counts, claim IDs, audit mode. Cheaper than aeo_fetch for context-window-constrained agents.

Inputs: origin (required).

aeo_well_known_urlA

Compute the canonical AEO well-known URL for an origin, without fetching.

Inputs: origin (required).

agent_card_inspectB

Structured summary of an Agent Card document. Pass EITHER url (the server fetches it) OR document_json (already-fetched).

Inputs: url, document_json.

agent_card_tool_disclosureB

Return the list of tools an agent declares, with side-effect class and (when present) MCP Tool Card URI for each tool.

Inputs: url, document_json.

agent_card_validateB

Validate an Agent Card JSON document against the v0.1 schema.

Inputs: document_json (required).

agent_card_well_known_urlA

Compute the canonical Agent Card well-known URL for a given origin + agent_id (convention: /.well-known/agents/<agent_id>.json).

Inputs: origin (required), agent_id (required).

ai_evidence_inspectC

Structured summary of an AI Evidence object: claim text, source metadata, retrieval method, synthesis role, hash, signed-or-not.

Inputs: document_json (required).

ai_evidence_validateC

Validate an AI Evidence object against the v0.1 schema.

Inputs: document_json (required).

ai_evidence_verify_hashA

Compute SHA-256 over the canonical UTF-8 form of candidate_text (LF endings, no trailing newline) and compare against the evidence's verification.content_hash. Returns ok=true on match, or { error: hash_mismatch, expected, recomputed }.

Inputs: document_json (required), candidate_text (required).

attestation_canonical_hashA

Compute the SHA-256 canonical-JSON hash of an arbitrary value (sorted keys, no whitespace). This is the structural hash convention used by procurement-decision-api, aeo-validator-service, aeo-graph-explorer-rs, and hash-attestation-rs. Identical JSON values produce identical hashes regardless of original whitespace or key order.

Inputs: body (required).

attestation_inspectA

Pretty-print an Attestation envelope with structural validation: confirms every required field is present, reports the decoded signature byte-length (should be 64), and surfaces the key_url + signed_at fields a caller would use to find the matching public key.

Inputs: attestation (required).

attestation_verifyA

Verify an ed25519 Attestation envelope against a body and caller-supplied public key. Recomputes the canonical hash and verifies the signature over the hash string. Does not verify that the supplied key belongs to the claimed key_url or signer. Public key accepted as 64-char hex or base64.

Inputs: body (required), public_key (required), attestation (required).

audit_chain_verifyA

Walk an array of GovernanceEvents top-to-bottom and verify the hash chain: monotonic event_id, prev_hash linkage, self-consistency of each event's hash. Returns { valid, checked, first_break_at, reason }, the same shape audit-stream-py's GET /verify endpoint emits.

Inputs: events (required).

audit_chain_verify_liveA

Ask the configured audit-stream service to check its full server-side hash chain. This checks chain continuity, not whether events are truthful, authorized, complete, or legally compliant. Returns valid, checked, first_break_at, and reason. Requires AUDIT_STREAM_URL and AUDIT_STREAM_TOKEN; returns a structured error otherwise.

Inputs: no named parameters.

audit_event_composeA

Build a ready-to-POST audit-stream-py GovernanceEvent: assigns event_id, computes the canonical hash, links to prev_hash (defaults to 64 zeros for event #1). Kind must be one of the 19 declared event kinds (decision_card_drafted, watch_drifted, request_denied, etc., plus 'other').

Inputs: kind (required), source (required), event_id (required), payload, +2 more.

audit_event_emitA

POSTs a governance event to the audit-stream-py instance configured by AUDIT_STREAM_URL and AUDIT_STREAM_TOKEN. This sends kind, caller-asserted source, and payload to that service. The bearer token authenticates this MCP process, not the claimed producer or tenant. Obtain the user's approval before recording an event; omit secrets and personal data unless the service is approved for them. The server assigns event_id/timestamp/prev_hash/hash. Returns a validated service event receipt or a structured error; receipt does not prove durable storage or event completeness.

Inputs: kind (required), source (required), payload.

audit_event_inspectA

Pretty-print one GovernanceEvent with structural validation: required fields, known/unknown kind, payload key list, and self-consistency check (does the event's hash match the recomputed canonical hash of the body?).

Inputs: event (required).

audit_events_queryA

GET recent governance events from a running audit-stream-py instance (AUDIT_STREAM_URL and AUDIT_STREAM_TOKEN), optionally filtered by kind or source. The service does not combine both filters. Returns the events array plus a count field; recorded source labels are caller-asserted. Returns a structured error when unconfigured.

Inputs: kind, limit, source.

aup_check_complianceA

Compare an AUP with a Student AI Disclosure on eight declared checks: policy effective window, signature, artifact_hash, teacher acknowledgment, prompt evidence mode, permitted/prohibited roles, assistance-extent ceiling, and assistance_extent_max=none vs ai_used=true. Returns { allowed, policy_id, disclosure_id, violations[] }; allowed=true means only these checks passed. It does not verify policy scope, tool/vendor requirements, parent consent, Tutor Cards, or legal compliance, and is not an enforcement decision.

Inputs: disclosure_json (required), aup_url, aup_json.

aup_fetchB

Fetch a Classroom AI AUP from a URL. Returns the parsed, schema-validated JSON document.

Inputs: url (required).

aup_inspectB

Structured summary of a Classroom AI AUP: policy identity, scope, permitted-use counts, prohibited-use counts, disclosure requirements, supervision level, vendor requirements posture, parent-consent gating.

Inputs: url, document_json.

aup_validateB

Validate a Classroom AI AUP JSON document against the v0.1 schema. Enforces conditional rules: course scope requires non-empty course_ids; assignment scope requires non-empty assignment_ids; assistance_extent_max=none forbids any permitted_roles; expires_at must follow effective_at; roles cannot be both permitted and prohibited.

Inputs: document_json (required).

aup_well_known_urlC

Compute the canonical Classroom AI AUP well-known URL: /.well-known/ai-aup.json.

Inputs: origin (required).

claims_card_chainA

Link a new AI Claims Decision Card to its predecessor: sets attestation.chain_index (prev + 1) and attestation.prev_card_hash. Validates the predecessor hash is 64-char hex and its index is a non-negative integer. Call claims_card_sign next.

Inputs: card (required), prev_card_hash (required), prev_chain_index (required).

claims_card_inspectB

Structured summary of an AI Claims Decision Card: claim type, outcome, coverage, evidence count, model, jurisdiction, human-in-loop, chain position, and attestation status. Cheaper than reading the full card.

Inputs: document (required).

claims_card_signA

Compute the canonical SHA-256 hash of an AI Claims Decision Card for ed25519 signing. Sorts object keys recursively and excludes attestation.card_hash + attestation.signature before hashing. Returns the 64-char lowercase hex card_hash; caller signs it offline.

Inputs: document (required).

claims_card_validateA

Validate an AI Claims Decision Card against the bundled v0.1 structural schema, including required fields, enums, ranges, and attestation field shapes. Returns the first schema issue or { valid, claims_card_id, version }. This does not verify signatures, evidence authenticity, the coverage decision, or legal compliance.

Inputs: document (required).

clinical_ai_fetchA

Fetch a Clinical AI Card from a URL. Returns the parsed, schema-validated JSON document.

Inputs: url (required).

clinical_ai_inspectC

Structured summary of a Clinical AI Card: system identity, clinical context (indication / care settings / patient population), regulatory posture (FDA / SaMD), clinical role, evidence (validation studies + bias audit + performance metrics), HIPAA + BAA posture, EHR integration (FHIR / SMART / CDS Hooks), safety + mandated reporting.

Inputs: url, document_json.

clinical_ai_validateA

Validate a Clinical AI Card JSON document against the v0.1 schema. Enforces the headline rules: autonomy ⇔ medical device, SaMD completeness, FDA-clearance documentation, PHI ⇒ explicit HIPAA + BAA posture, and bias_audit_uri requirement for SaMD class II+ / autonomous / pre-authorization use.

Inputs: document_json (required).

clinical_ai_well_known_urlB

Compute the canonical Clinical AI Card well-known URL: /.well-known/clinical-ai/<system_id>.json.

Inputs: origin (required), system_id (required).

decision_card_fetchA

Fetch an AI Procurement Decision Card from a URL. Returns the parsed, schema-validated JSON document.

Inputs: url (required).

decision_card_infer_statusA

Validate each rubric row, then infer a preview decision.status. Mirrors procurement-decision-api's rubric engine: any 'fail' -> 'rejected-with-remediation'; any 'partial' or 'pass-with-condition' -> 'approved-with-conditions'; all 'pass' -> 'approved'; empty or all 'n/a' -> 'pending'. An unsupported result returns an error, never an approval.

Inputs: rubric (required).

decision_card_inspectC

Structured summary of an AI Procurement Decision Card: buyer identity, decision status + scope, vendor + documents reviewed (by type and URL), rubric pass/partial/fail counts, conditions count, signatures count, publication posture, history event count, withdrawal flag.

Inputs: url, document_json.

decision_card_signature_checkA

Structural check on a Decision Card's signatures[] block: count signers, show their method/key/timestamp, and return the canonical-JSON hash of the card body (excluding signatures) so a caller can pair this with attestation_verify for cryptographic checking.

Inputs: url, document_json.

decision_card_to_policy_bundleA

Read-only deny preview for rejected, withdrawn, expired, or pending Decision Cards; returned bundles carry preview_only=true and enforceable=false. Approved statuses return an MCP error: only the live policy engine can check buyer identity and trusted attestation, vendor scope, allowed action, and effective window before producing an enforceable bundle. Do not use this tool output as an access grant.

Inputs: url, document_json.

decision_card_validateA

Validate an AI Procurement Decision Card JSON document against the v0.1 schema. Enforces conditional rules: status=approved-with-conditions and status=rejected-with-remediation require at least one entry in conditions; status=withdrawn requires a withdrawal block; publication.is_public=true requires publication_uri.

Inputs: document_json (required).

decision_card_well_known_urlB

Compute the canonical AI Procurement Decision Card well-known URL: /.well-known/decisions/<decision_id>.json.

Inputs: origin (required), decision_id (required).

defensetech_audit_event_check_invariantsA

Run three deterministic DefenseTech event checks: CUI distribution-statement presence, ITAR status with tokenized license for an authorized foreign person, and a 72-hour discovery-to-filing interval. The last check requires event.discovered_at; event.timestamp is not a substitute. Returns ok + errors + passed, not a legal compliance determination.

Inputs: event (required).

defensetech_check_cui_distribution_statementB

Check that a CUI-Specified+ tier event carries the required distribution_statement (DoDI 5230.24). For PUBLIC / CUI-BASIC tiers reports applicable=false.

Inputs: cui_categorization (required), distribution_statement.

defensetech_check_dfars_72h_clockA

Calculate the DFARS 252.204-7012 discovery-to-filing 72-hour interval from supplied timezone-qualified timestamps. Requires discovered_at, not occurred_at. Invalid or pre-discovery filing timestamps return valid=false and within_window=false. This arithmetic preview does not establish a reporting duty or prove filing.

Inputs: filed_at (required), discovered_at (required).

defensetech_check_itar_us_personB

Check that an ITAR resource event has US-PERSON-VERIFIED (or AUTHORIZED-FOREIGN-PERSON-WITH-LICENSE + DDTC license number tokenized) on the agent. Per 22 CFR 120.62 / 22 CFR 120.50 deemed-export rule.

Inputs: export_control_status (required), human_user_us_person_status, ddtc_export_license_number_tokenized.

defensetech_incident_classify_event_typeA

Given a freeform description of a defense-AI incident, classify it into one of the 22 DefenseTech Incident Card event_type values (DFARS cyber / CUI spillage / ITAR violation / foreign-person breach / classified misuse / NISPOM insider-threat / CMMC POA&M / AI-tool supply-chain / etc.). Returns top 3 candidates with token-match scoring.

Inputs: description (required).

defensetech_summarize_cmmc_evidence_bundleA

Summarize a CMMC L2/L3 readiness evidence bundle: target level, assessment mode, evidence count, family coverage count, outcome distribution, orphan-failure count (not-satisfied without poam_ref), SPRS evidence presence, and quick invariant checks for POA&M traceability + SPRS-when-7019/7020-in-scope.

Inputs: bundle (required).

defensetech_vault_contract_cross_binding_checkA

Verify the cross_binding_refs block on a DefenseTech vault contract is syntactically valid: all referenced repos are HTTPS URLs. Returns valid_refs + errors. Does NOT fetch remotely (syntactic check only).

Inputs: contract (required).

defensetech_vault_resolve_3axisB

Deterministic preview of a DefenseTech 3-axis vault contract: intersect allowed_actions and OR requires_* flags. Unknown or incomparable human-user requirements return ok=false for an authorized policy owner to resolve. This does not verify identity, classification, licenses, or enforce access.

Inputs: tuple (required), contract (required).

disclosure_aup_checkA

Surface the disclosure's declared policy posture and any aup_uri reference. Status is one of: declared_compliant, declared_non_compliant, aup_referenced_but_unclaimed, no_aup_reference. This does not verify compliance; aup_check_compliance evaluates eight declared AUP/Disclosure checks only.

Inputs: document_json (required).

disclosure_inspectC

Structured summary of a Student AI Disclosure: assignment identity, AI usage facts, tools used (with back-refs to agent / tutor cards), role taxonomy, assistance extent, prompt-mode + count, artifact hash, policy posture, signature + acknowledgment.

Inputs: document_json (required).

disclosure_validateA

Validate a Student AI Disclosure JSON document against the v0.1 schema. Enforces conditional rules (ai_used true requires tools/roles/extent/prompt_mode; ai_used false forbids them; prompt mode gates prompts presence).

Inputs: document_json (required).

disclosure_verify_artifact_hashA

Recompute SHA-256 over a candidate artifact and compare to disclosure.artifact_hash. Pass either candidate_text (canonical SHA-256: LF, no trailing newline) for text artifacts, or candidate_bytes_base64 (raw-bytes SHA-256) for binary artifacts like PDFs / images. Returns ok=true on match, otherwise { error: artifact_hash_mismatch, expected, recomputed }.

Inputs: document_json (required), candidate_text, candidate_bytes_base64.

disclosure_verify_prompt_hashA

Verify a single prompt hash in a hashed-mode disclosure. Looks up prompt_id and compares canonical SHA-256 of candidate_text. Returns ok=true on match or { error: prompt_hash_mismatch, expected, recomputed }. Errors with wrong_prompt_mode if prompt_evidence_mode is not 'hashed'.

Inputs: prompt_id (required), document_json (required), candidate_text (required).

incident_affected_walkA

Walk an Incident Card's affected block and return every referenced Suite document as { uri, kind }. Useful as the seed list for incident-correlation-rs or fan-out validation via aeo-validator-service.

Inputs: url, document_json.

incident_fetchA

Fetch an AI Incident Card from a URL. Returns the parsed, schema-validated JSON document.

Inputs: url (required).

incident_index_fetchA

HEADLINE TOOL, fetch a vendor's /.well-known/ai-incidents.json index and return a procurement-friendly summary: total count, breakdown by severity, breakdown by status, IDs sorted by disclosed_at descending. The cheapest way for a CISO or procurement reviewer to scan a vendor's incident history.

Inputs: origin (required).

incident_inspectC

Structured summary of an AI Incident Card: incident identity (id, title, severity, categories, status, timeline), affected products + versions + Agent/Tutor/Tool Card cross-references, root cause, harm, mitigation, regulatory filings, withdrawal posture, revision metadata.

Inputs: url, document_json.

incident_remediation_planC

Map each affected URI in an Incident Card to a recommended Action + Urgency. Single-hop preview of what incident-correlation-rs.correlate() would produce: agent/tutor/tool-card -> revalidate; vendor/product -> request_review. Urgency follows severity.

Inputs: url, document_json.

incident_validateA

Validate an AI Incident Card JSON document against the v0.1 schema. Enforces conditional rules: status=resolved requires resolved_at; status=withdrawn requires withdrawal block; non-empty regulatory.reported_to requires non-empty regulatory_filing_uris; root_cause.category=other and categories containing 'other' both require the corresponding _other_text fields.

Inputs: document_json (required).

incident_well_known_urlA

Compute the canonical AI Incident Card well-known URL: /.well-known/ai-incidents/.json.

Inputs: origin (required), incident_id (required).

prompt_provenance_eval_resultC

Extract a single evaluation suite's result from a Prompt Provenance document.

Inputs: suite_name (required), document_json (required).

prompt_provenance_inspectC

Structured summary of a Prompt Provenance document: prompt identity, lineage, approval state, evaluation suites.

Inputs: document_json (required).

prompt_provenance_validateA

Validate a Prompt Provenance JSON document against the v0.1 schema. Returns { valid, prompt_id, version } or { valid: false, reason }.

Inputs: document_json (required).

suite_doc_detect_specA

Detect which Kinetic Gain Suite spec a JSON document is by sniffing its top-level *_version field. Returns { spec, version_field, version }. Recognises all 12 specs; returns spec='unknown' for anything else.

Inputs: body (required).

suite_doc_driftB

Structural diff between two versions of the same Suite document. Returns { drifted, spec_before, spec_after, spec_changed, content_hash_before, content_hash_after, added_fields, removed_fields, changed_fields }, mirrors the DriftReport shape aeo-validator-service emits for watch rechecks.

Inputs: after (required), before (required).

tool_card_inspectC

Structured summary of an MCP Tool Card: tool identity, safety profile (side-effect class, PII/secrets exposure, approval requirement), test count, p99 latency.

Inputs: url, document_json.

tool_card_tested_withB

Return the tested-LLM entries for a tool, optionally filtered by a substring match on the LLM identifier.

Inputs: url, llm_filter, document_json.

tool_card_validateC

Validate an MCP Tool Card JSON document against the v0.1 schema.

Inputs: document_json (required).

tool_card_well_known_urlA

Compute the canonical MCP Tool Card well-known URL (convention: /.well-known/mcp-tools/<tool_name>.json).

Inputs: tool_name (required), mcp_server_origin (required).

tutor_card_coppa_checkB

Enforce the spec's COPPA conditional rule: if audience.age_range_min < 13, data_privacy.coppa_compliant MUST be true. Returns { ok: true } or { error: coppa_violation, reason }.

Inputs: url, document_json.

tutor_card_fetchB

Fetch a Tutor Card from a URL. Returns the parsed, schema-validated JSON.

Inputs: url (required).

tutor_card_inspectC

Structured summary of a Tutor Card: tutor identity, audience, pedagogy, subject scope counts, safety strength, FERPA/COPPA/GDPR posture, evaluation count, COPPA-rule check.

Inputs: url, document_json.

tutor_card_subject_checkC

Classify a topic against the tutor's subject scope. Returns one of: primary, included, excluded, unknown, with the matched term when applicable.

Inputs: query (required), url, document_json.

tutor_card_validateA

Validate an AI Tutor Card JSON document against the v0.1 schema, including the COPPA conditional rule (age_range_min < 13 ⇒ coppa_compliant must be true).

Inputs: document_json (required).

tutor_card_well_known_urlB

Compute the canonical AI Tutor Card well-known URL (convention: /.well-known/tutors/<tutor_id>.json).

Inputs: origin (required), tutor_id (required).

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3/5.0

    Scored across 75 tools

    Disambiguation3/5

    The tool set has a clear spec-prefix convention (e.g., clinical_ai_validate, tutor_card_inspect), so many tools are distinguishable. However, 75 tools repeat the same validate/inspect/fetch/check actions across a dozen similar document schemas, so an agent can easily select the wrong spec's validator or inspector without careful reading.

    Naming Consistency4/5

    Names are overwhelmingly snake_case with a consistent subject-prefix + action pattern (ai_evidence_validate, tutor_card_coppa_check, audit_chain_verify). Minor deviations exist, such as aeo_get_claim using 'get' instead of fetch/inspect, audit_events_query using a plural subject, and defensetech_* tools adding an extra domain prefix, but the overall convention is readable and stable.

    Tool Count1/5

    At 75 tools, this server is far beyond the 3-15 range appropriate for a single MCP server, and exceeds the 50+ threshold for extreme mismatch. The functionality spans many independent specs and services that would be better split into separate servers.

    Completeness3/5

    The surface covers validation, inspection, fetching, hashing, and compliance checks for many Suite specs, but coverage is uneven: AEO lacks a validate tool, several specs lack fetch tools, and signing/hashing support is inconsistent (e.g., claims_card_sign exists but no equivalent sign tool for most other cards). There are also no update/delete operations, though immutability may justify that for attestation documents.

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI agents to operate through a local, signed boundary that blocks prompt injection and secret leakage, verifies outputs, preserves cross-session memory, and provides offline-verifiable receipts. It also exposes 900+ MCP tools for discoverable agent actions.
      7
      MIT
    • F
      license
      B
      quality
      B
      maintenance
      Enables AI agents to query an air-gapped enterprise knowledge vault via seven MCP tools for deterministic and hybrid retrieval, entity graph traversal, context compression, archive inspection, onboarding scans, and node status.
      7
      -
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables autonomous agents to securely access and manage virtual file systems, records, schemas, SQL and Mongo queries, and RAG search through scoped MCP tools, while supporting binary transfers via presigned S3 recipes.
      PolyForm Noncommercial 1.0.0