{"full_name":"pipelab-org/pipelock","owner":"pipelab-org","name":"pipelock","shard":"1","url":"https://github.com/pipelab-org/pipelock","description":"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.","homepage":"https://pipelab.org","topics":["ai-agents","ai-security","dlp","golang","llm-security","security","ssrf-protection","egress-proxy","integrity-monitoring","mcp","github-action","security-scanning","agent-security","ai-agent-security","ai-firewall"],"tag_pages":["agent-security","ai-agents","ai-security","golang","llm-security","mcp","security"],"language":"Go","license":"Apache-2.0","avatar":"https://avatars.githubusercontent.com/u/340203353?v=4","created_at":"2026-02-08T00:40:18Z","pushed_at":"2026-11-16T04:49:05Z","metrics":{"stars":996,"forks":113,"watchers":4,"issues":67,"prs":2138,"releases":53,"issues_open":3,"issues_closed":64,"prs_open":1,"prs_closed":80,"prs_merged":2057,"discussions":12},"growth":{"day":{"stars":3,"forks":0,"base_day":"2026-11-15"}},"history":{"dates":["2026-11-12","2026-11-13","2026-11-14","2026-11-15","2026-11-16"],"stars":[990,990,991,993,996],"forks":[113,113,114,113,113],"watchers":[5,5,5,4,4],"issues_open":[3,3,3,3,3],"prs_open":[6,4,1,0,1]},"boards":{},"same_owner":[],"issues":[{"number":1308,"title":"Secret-egress evidence and coverage map","comments":13,"created_at":"2026-08-22T00:36:42Z"},{"number":1793,"title":"Extension idea: external pre-action evidence hook before web/MCP ingestion","comments":3,"created_at":"2026-10-04T18:57:32Z"},{"number":590,"title":"mcp: track 2026-07-28 RC runtime-security fixtures","comments":2,"created_at":"2026-05-23T16:28:20Z"}],"discussions":[{"number":2189,"title":"Request for review: egress-control benchmark of Pipelock presets, before I publish","comments":1,"category":"General","created_at":"2026-11-11T20:59:49Z"},{"number":1706,"title":"Research question: what worries people most when agents act on real systems?","comments":1,"category":"General","created_at":"2026-09-27T16:38:07Z"},{"number":1347,"title":"What is the Completeness grade computed from?","comments":1,"category":"Q&A","created_at":"2026-08-25T10:44:19Z"},{"number":778,"title":"Comparing notes: signed egress receipts and per-call tool-action decisions","comments":1,"category":"General","created_at":"2026-06-15T07:31:55Z"},{"number":333,"title":"v2.1.0: prove what your agent actually did","comments":0,"category":"Announcements","created_at":"2026-03-31T02:32:34Z"},{"number":244,"title":"v1.3.0 shipped + Pro tier is live","comments":0,"category":"Announcements","created_at":"2026-03-16T11:06:18Z"},{"number":86,"title":"v0.2.0: Bidirectional MCP Scanning, Project Audit, and 1,000+ Tests","comments":0,"category":"Announcements","created_at":"2026-02-14T00:05:40Z"},{"number":65,"title":"What AI agents are you running Pipelock with?","comments":0,"category":"General","created_at":"2026-02-09T23:02:02Z"}],"latest_release":{"name":"v3.8.0","tag":"v3.8.0","published_at":"2026-11-10T02:08:35Z","body":"\n### Breaking Changes / Upgrade Notes\n\n- **Metrics move off the agent-facing proxy port.** `/metrics` and `/stats` are available only when `metrics_listen` names a dedicated listener. Omitted, null, or blank values expose neither route at startup. Managed containment rejects a reload that removes or blanks the setting. The existing listener stays bound, `/metrics` returns `403` until a valid reload clears the denial, and `/stats` keeps serving loopback clients. Restart to disable the listener. Update scraper targets and keep agents off the unauthenticated metrics listener with network controls. Changing the address requires restart. [Upgrade guide](docs/guides/upgrading-to-3.8.md). ([#2044](https://github.com/pipelab-org/pipelock/pull/2044))\n- **New container and chart releases use `ghcr.io/pipelab-org`.** Update image repositories, registry policies and Helm OCI URLs when adopting this release. Releases through v3.7.1 retain their original registry addresses. v3.7.1 attestations use `pipelab-org/pipelock`; releases through v3.7.0 retain the `luckyPipewrench/pipelock` signing identity. Generated Compose and sidecar output also adopt the new image defaults. The Go module path and Homebrew tap stay unchanged. [Verified installation](README.md#verify-release-integrity). ([#1911](https://github.com/pipelab-org/pipelock/pull/1911), [#2020](https://github.com/pipelab-org/pipelock/pull/2020), [#2062](https://github.com/pipelab-org/pipelock/pull/2062))\n- **Expired temporary scanner exceptions lapse instead of preventing startup.** A valid expired entry emits a startup or reload warning and ordinary scanning resumes for its scope. Malformed entries still reject configuration. Running scanners enforce expiry without waiting for reload; there's no separate clock-triggered warning. ([#1942](https://github.com/pipelab-org/pipelock/pull/1942))\n- **Non-finite floating-point settings are rejected.** Replace NaN or infinity values with finite values inside the setting's supported range before upgrading. The configuration error names the affected field. ([#2029](https://github.com/pipelab-org/pipelock/pull/2029))\n- **The default URL ceiling increases to 8000 octets.** Explicit limits still win, and presets with tighter limits retain them. Operators who require the previous ceiling should set `fetch_proxy.monitoring.max_url_length` explicitly. Longer URLs can carry more data per request; the other URL checks still apply. ([#2052](https://github.com/pipelab-org/pipelock/pull/2052))\n- **Generated audit configurations retain explicit tool-policy blocks and disable adaptive escalation.** Ordinary findings remain warnings. Regenerated strict configurations also retain their bundled actions, taint policy and tool-chain detection. Review regenerated policy before replacing a custom configuration. ([#1943](https://github.com/pipelab-org/pipelock/pull/1943), [#1961](https://github.com/pipelab-org/pipelock/pull/1961))\n- **Regenerated agent presets and the quickstart retain built-in and supplemental response patterns.** The `claude-code`, `cursor`, `generic-agent`, and `hostile-model` presets give supplemental expressions `Preset` names. Review name-based alerts and suppressions when replacing older YAML; existing saved configurations keep their names until edited or replaced. [Preset migration](docs/configuration.md#presets). ([#2063](https://github.com/pipelab-org/pipelock/pull/2063), [#2066](https://github.com/pipelab-org/pipelock/pull/2066))\n- **Core findings remain visible in repository scans.** `git scan-diff`, `audit`, and the GitHub Action apply the core DLP floor even with custom patterns or defaults disabled. Inline markers and rule suppressions can't silence core findings; path exclusions still exclude the entire path. Review synthetic fixtures and narrowly scoped exclusions before upgrading. [Suppression guide](docs/guides/suppression.md). ([#2064](https://github.com/pipelab-org/pipelock/pull/2064))\n- **Codex MCP setup forwards environment vari","is_prerelease":false},"readme":true,"readme_zh":false,"news":[],"stale":false,"snapshot_at":"2026-11-16T09:23:14Z","history_start":"2026-09-06"}