MCP Kinetic Gain
@mizcausevic-dev
MCP Kinetic Gain について
Unified MCP server for all 11 Kinetic Gain Protocol Suite specs + DefenseTech module (8 tools) — 71 callable tools (47 spec + 16 implementation + 8 DefenseTech). One Claude Desktop config entry. Stdio, no API key, no build step.
基本情報
設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"kinetic-gain": {
"command": "npx",
"args": [
"-y",
"mcp-kinetic-gain"
]
}
}
}ツール
50Fetch the full AEO Protocol declaration at an origin's
Return a structured summary of an AEO declaration: entity
Extract a single AEO claim by ID
Compute the canonical AEO well-known URL for an origin
Validate a Prompt Provenance JSON document against the v0.1
Structured summary of a Prompt Provenance document: prompt
Extract a single evaluation suite's result from a Prompt
Compute the canonical Agent Card well-known URL for a given
Structured summary of an Agent Card document
Return the list of tools an agent declares, with side-effect
Validate an Agent Card JSON document against the v0.1 schema.
Validate an AI Evidence object against the v0.1 schema.
Structured summary of an AI Evidence object: claim text
Compute SHA-256 over the canonical UTF-8 form of
Compute the canonical MCP Tool Card well-known URL
Structured summary of an MCP Tool Card: tool identity, safety
Return the tested-LLM entries for a tool, optionally filtered
Validate an MCP Tool Card JSON document against the v0.1
Compute the canonical AI Tutor Card well-known URL
Fetch a Tutor Card from a URL
Validate an AI Tutor Card JSON document against the v0.1
Structured summary of a Tutor Card: tutor identity, audience
Classify a topic against the tutor's subject scope
Enforce the spec's COPPA conditional rule: if
Validate a Student AI Disclosure JSON document against the
Structured summary of a Student AI Disclosure: assignment
Recompute SHA-256 over a candidate artifact and compare to
Verify a single prompt hash in a hashed-mode disclosure
Surface the disclosure's policy posture: whether an aup_uri
Compute the canonical Classroom AI AUP well-known URL
Fetch a Classroom AI AUP from a URL
Validate a Classroom AI AUP JSON document against the v0.1
Structured summary of a Classroom AI AUP: policy identity
HEADLINE TOOL, joins an AUP with a Student AI Disclosure and
Compute the canonical Clinical AI Card well-known URL
Fetch a Clinical AI Card from a URL
Validate a Clinical AI Card JSON document against the v0.1
Structured summary of a Clinical AI Card: system identity
Compute the canonical AI Incident Card well-known URL
Fetch an AI Incident Card from a URL
Validate an AI Incident Card JSON document against the v0.1
Structured summary of an AI Incident Card: incident identity
HEADLINE TOOL, fetch a vendor's
Compute the canonical AI Procurement Decision Card well-known
Fetch an AI Procurement Decision Card from a URL
Validate an AI Procurement Decision Card JSON document
Structured summary of an AI Procurement Decision Card: buyer
Given a rubric, infer the right decision.status
Translate a Decision Card into the PolicyBundle that
Structural check on a Decision Card's signatures[] block
概要
What is MCP Kinetic Gain?
MCP Kinetic Gain is a single MCP server that exposes 75 tools covering all twelve specs of the Kinetic Gain Protocol Suite plus DefenseTech implementation tooling. Designed for agents in MCP-compatible clients (Claude Desktop, Cursor, etc.), it enables validation, inspection, and cross-spec workflows for governance and compliance documents.
How to use MCP Kinetic Gain?
Install globally via npm install -g mcp-kinetic-gain or run without installing via npx mcp-kinetic-gain. Add a single entry to your client’s MCP server config (e.g., claude_desktop_config.json) with command npx -y mcp-kinetic-gain. The server also doubles as a CLI validator for Suite files using npx mcp-kinetic-gain validate <path>.
Key features of MCP Kinetic Gain
- 75 tools across 12 Kinetic Gain Suite specs
- Hash attestation (ed25519) and audit-stream chain verification
- Cross-spec drift detection between document versions
- DefenseTech vault resolver and invariant checkers
- AI Claims Decision Card (InsurTech) support
- CLI mode for CI/pre-commit validation
Use cases of MCP Kinetic Gain
- Validate an AEO declaration and inspect entity claims
- Check a Student AI Disclosure against its classroom AI AUP for compliance
- Verify an AI Evidence hash against a candidate text
- Fetch and inspect an Agent Card or Tool Card via its well-known URL
- Audit an incident chain by walking events and verifying signatures
FAQ from MCP Kinetic Gain
Why one server instead of five separate ones?
A single config entry avoids managing multiple servers, enables atomic cross-spec workflows, shares schemas, and provides a deprecation path for predecessor servers.
How do I install the server?
Install globally via npm install -g mcp-kinetic-gain or run ephemerally with npx mcp-kinetic-gain. The README also shows a Claude Desktop config using npx -y.
Which specs are supported?
All twelve: 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, and AI Claims Decision Card.
Can I use the tools outside an MCP client?
Yes. The same binary offers a CLI mode: npx mcp-kinetic-gain validate <file> auto-detects the spec and validates against the zod schemas. It also supports globs and GitHub Actions annotations.
What is the license?
The server code is AGPL-3.0 (reference implementation; commercial SaaS hosts must share modifications). The specs themselves are MIT.
「AI とエージェント」の他のコンテンツ
LinkedIn MCP Server
stickerdanielOpen-source MCP server for LinkedIn. Give Claude and any MCP-compatible AI agent access to profiles, companies, jobs, and messages.
MCP-NixOS - Because Your AI Assistant Shouldn't Hallucinate About Packages
utensilsMCP-NixOS - Model Context Protocol Server for NixOS resources
MCP Server - Remote MacOs Use
baryhuangThe only general AI agent that does NOT requires extra API key, giving you full control on your local and remote MacOs from Claude Desktop App
meGPT - upload an author's content into an LLM
adriancoCode to process many kinds of content by an author into an MCP server
Open Multi-Agent Canvas
CopilotKitThe open-source multi-agent chat interface that lets you manage multiple agents in one dynamic conversation and add MCP servers for deep research
コメント