Guardrailed FHIR access for AI agents: PHI redaction, audit trail, step-up auth, tenant isolation
FinishKit MCP: scan GitHub repos for security vulnerabilities, deployment blockers, and quality
Connects AI assistants to IDA Pro for decompilation, disassembly, database changes, and reverse-engineering workflows.
Key Features: IDA Pro decompilation access Database comments, renaming, and type updates Enforces YAML permissions on agent tool calls, inspects sensitive data and tool chains, and records audit trails.
Key Features: PII and PHI classification Dangerous tool-chain detection Remote MCP server for querying RAD Security findings, reports, and runtime data across Kubernetes and cloud environments.
Key Features: Hosted Streamable HTTP MCP endpoint Security findings queries Security report retrieval Binary Ninja plugin and MCP server enabling AI-assisted binary analysis and reverse engineering via MCP clients.
Key Features: Real-time integration between Binary Ninja and MCP clients Support for multiple MCP clients including Cline, Claude Desktop, Roo Code, Cursor, Windsurf, Claud… Automatic switching between multiple open binaries Issues signed agent passports, manages scoped delegation, and enforces governance through identity, policy, gateway, and coordination tools.
Key Features: Ed25519 agent identity and passport issuance Cryptographic delegation verification and revocation Policy-enforcing gateway workflows Creates and reads AES-256-GCM encrypted, one-time Volta Notes for credential handoff between users and AI agents.
Key Features: Create AES-256-GCM encrypted notes Read and permanently destroy notes Keep decryption keys in URL fragments Governance proxy that filters MCP tool results for prompt injection and capability abuse using configurable policy modes.
Key Features: MCP client-to-server governance proxy Prompt injection filtering before results reach agents Capability abuse prevention Proxies MCP traffic to add signed audits, redaction, policies, rate limits, and a local read-only dashboard.
Key Features: Stdio and HTTP MCP proxying JSONL and SQLite audit storage HMAC-SHA256 audit signing Requires named human approval for consequential agent actions and issues verifiable Ed25519 trust receipts.
Key Features: Named human approval gates Offline receipt verification UI/UX design-audit MCP server: scores a project on 12 dimensions vs WCAG 2.2 + APCA.
