Read-only MCP server for coding-agent traces, alerts, costs, usage, and budget health.
Read-only MCP (Model Context Protocol) server that connects coding agents to the AgentGuard Read API. Use it after the local SDK is already in place and you want Codex, Claude Code, Cursor, or another MCP client to inspect traces, decision events, alerts, usage, costs, and budget health. The boundary is deliberate: The SDK proves runtime enforcement locally The MCP server gives coding agents a…
Verification confirms publisher identity (repo ownership), not code safety. The security scan covers known CVEs and suspicious install scripts — it cannot prove the absence of malicious code.
Read-only MCP (Model Context Protocol) server that connects coding agents to the AgentGuard Read API. Use it after the local SDK is already in place and you want Codex, Claude Code, Cursor, or another MCP client to inspect traces, decision events, alerts, usage, costs, and budget health. The boundary is deliberate: The SDK proves runtime enforcement locally The MCP server gives coding agents a narrow read surface over retained data The hosted dashboard stays the operational control plane Tool |…