Security

Guardian layer for AI agents: identity, secrets, audit via MCP.

Delegated OAuth 2.1 bidding and creative tools for one advertiser.

Prediction markets, AI-built index funds, event cards and a CLOB exchange as MCP tools. 71 read tools work anonymously; trading needs the user's own API key or an OAuth 2.1 token with the trade scope. Withdrawals never happen through this connector.

Identity infrastructure. Read a person's verified record over MCP. When a paid query runs against it, 75% of the payment goes to that person.

Read-only tools and resources for SendHQ's public API, authentication, pricing, and integration documentation.

Figma designs pulled once into a local vault; AI agents read them offline with no rate limits

Local-first agent persona runtime (MCP): cross-session personality that survives compaction.

Routes each task to the few skills, MCP servers and tools that fit it.

Wraps the ShimGuard CLI as a single generic MCP tool for agent-shim provenance checks.

Provenance gateway for stdio MCP servers: blocks tool calls fed by untrusted tool results.

MCP server for VerusID identities — register, update, revoke, recover, query, set timelock

Encrypted secrets for Next.js. AES-256-GCM with no vault needed.

MCP auth gateway: verify agent identity, enforce per-tool permissions, block replays, emit receipts.

Scan local or remote hosts for open TCP ports. Runs locally.

Authorization layer between AI agents and money: ALLOW/APPROVAL/DENY, budgets, audit.

OAuth 2.1/OIDC resource-server reference for secure MCP authorization

AI agent security: prompt injection detection, semantic memory, output scanning, prompt hardening

Default trust layer for agentic payments: authorize agent spend, evidence, receipts, settlement.

Ybe Check – security and compliance orchestrator for AI-generated repositories.

Policy-based governance for AI agent tool calls. YAML policy, approval gates, audit logging.