pipelock

View on GitHub

Open-source AI agent firewall for MCP security and agent egress. Scans mediated HTTP, MCP, A2A, and WebSocket traffic for exfiltration, SSRF, and prompt injection, and emits mediator-signed action receipts: verifiable audit evidence from outside the agent.

★ 912104 forksGoApache-2.0luckyPipewrench

Go-based AI agent firewall that proxies and inspects HTTP, MCP, A2A, and WebSocket traffic for secret exfiltration, prompt injection, and SSRF, then emits mediator-signed action receipts for offline verifiable audit. Ships GitHub Action, sandbox and cluster deployment options.

Use Cases

Block secret exfiltration from AI agent egressScan mediated HTTP/MCP/A2A/WebSocket trafficDetect prompt injection and tool poisoningPrevent SSRF and cloud metadata endpoint accessEmit mediator-signed action receipts as audit evidenceVerify signed receipts offline for complianceGateway for MCP server traffic securityRun as a GitHub Action in CI security pipelineEnforce per-agent egress policy and budgetsOWASP LLM/Agentic Top 10 mitigation mapping

Built With

Language: Go
Frameworks: Claude Code · OpenAI Codex · Cline · OpenCode · Cursor · Zed · VS Code · JetBrains · OpenAI Agents SDK · Google ADK · AutoGen · CrewAI · LangGraph · Cobra · Prometheus · OpenTelemetry

pipelock

Use Cases

Built With

Tags