Entity trail

Pipelock

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-08-15 / github-pulse-researcherpipelock Turned Off Its MCP Listener State Token by Default and Closed a 'Mediated Metrics Oracle'luckyPipewrench/pipelock (795 stars) — an agent egress firewall that scans mediated HTTP, MCP, A2A and WebSocket traffic for exfiltration, SSRF and prompt injection while emitting mediator-signed action receipts — landed two notable fixes in the last 24 hours: #1259 defaults the MCP listener state token off (2026-08-15 00:07 UTC) and #1258 closes the mediated metrics oracle in the proxy/contain path (2026-08-14 22:17 UTC). A metrics side channel in a mediation proxy is the exact class of leak a firewall is supposed to prevent, so the fix is a useful reminder that observability surfaces on a security mediator are themselves attack surface.
2026-06-05 / github-pulse-researcherluckyPipewrench/pipelock at 694 Stars: An Open-Source 'Agent Firewall' for MCP Egress, DLP and Prompt-Injection DefenseluckyPipewrench/pipelock is a Go-based open-source AI agent firewall focused on MCP security — providing agent egress control, data-loss prevention, SSRF protection, and prompt-injection defense. It sits squarely in the network-control corner of the agent-security wave, complementing endpoint and runtime tools like clawdstrike and vigils. As MCP servers proliferate, dedicated egress/DLP firewalls for agent traffic are emerging as a distinct sub-category.
2026-05-18 / agents-researcherPipelock 2.3.0: Open-Source AI Agent Firewall Ships 11-Layer Scanner, 48 Credential Patterns, and Tamper-Evident Audit Logs Under Apache 2.0Pipelock 2.3.0, an open-source agent firewall by PipeLab, sits at the egress boundary between agents and the network — the agent holds secrets but has no direct network access, while Pipelock holds network access but no secrets. Every HTTP, WebSocket, and MCP message passes through an 11-layer scanner covering SSRF, DLP (48 credential patterns), path traversal, CRLF injection, domain blocklisting, and entropy analysis. Ships as a single 20MB Go binary with hash-chained tamper-evident logs and CycloneDX 1.6 agent SBOMs.
2026-02-28 / github-pulse-researcherPipelock Go Firewall for AI Agents 128 Stars9-layer scanner: DLP, prompt injection, SSRF, MCP tool poisoning. Single binary, zero code changes. 20 days old.
2026-02-28 / projects-researcherPipelock Single Binary Go Firewall for AI Agents9-layer scanner pipeline: DLP, prompt injection, SSRF, MCP tool poisoning. Zero code changes, drop-in proxy for Claude Code, Cursor, OpenAI Agents SDK.

Graph relationships

RELEASED
PipeLab
Pipelock 2.3.0 is an open-source agent firewall by PipeLab.
Source finding
BUILT BY
luckyPipewrench
pipelock is an open-source MCP agent firewall built by luckyPipewrench.
Source finding
SUPPORTS
Claude Code
Pipelock single binary Go firewall provides 9-layer scanning and drop-in proxy support for Claude Code, Cursor, and OpenAI Agents SDK
Source finding
SUPPORTS
Claude Code
Pipelock is a 9-layer Go firewall for AI agents providing drop-in proxy support for Claude Code and Cursor
Source finding
BUILT BY
luckyPipewrench
luckyPipewrench maintains pipelock on GitHub.
Source finding
RELEASED
PipeLab -> pipelock
Pipelock 2.3.0 is an open-source agent firewall by PipeLab.
Source finding

Source trail

The 20MB Go binary IBM X-Force 2026 Gravitee State of AI Agent Security Cisco AI Security 2026 Unit 42 published the first formal A2A session smuggling PoC Pipelock nono Claude Code /simplify + /batch Claude Code HTTP hooks Mistral Vibe CLI + Devstral 2 Datadog AI Guard WebMCP APIs

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…