KeygraphHQ /shannon
Shannon is an AI pentester for web applications and APIs. It analyzes your source code, identifies attack vectors, and executes real exploits to prove vulnerabilities before they reach production.
★ 51,921⑂ 6,072👁 235⑃ 309● 175▤ 30
+31
日增长
+430
周增长
51,921
stars
6,072
forks
235
watchers
309
prs
175
issues
30
releases
README
[!NOTE] php2shell upgrade verification: confirm your PHP apps are patched against the php2shell RCE chain with the experimental --exp-php2shell probe. See the test results. Shannon is an autonomous, AI pentester for web applications and APIs. It analyzes your source code, identifies attack paths, and executes real exploits to prove vulnerabilities before they reach production. No exploit, no report. This repository is Shannon: the full open-source agent, run locally from your command line. Launch Shannon bash npx @keygraph/shannon@latest The interactive launcher will guide you through setup and your first pentest. --- --- [!TIP] AI agents and LLMs: start with llms.txt for a concise map of this repository, or use llms-full.txt for the README and docs combined into one file. Table of Contents - Table of Contents - What is Shannon? - Why Shannon Exists - Why "Shannon"? - Not a replacement for human pentesters - Shannon in Action - Quick Start - Prerequisites - Run Shannon - Key Capabilities - CI/CD Integrations - GitHub Actions - Editions - Architecture - Documentation - Safety, Scope, and Limitations - License - Acknowledgements - About Keygraph - Community and Support - Common Questions - Can I self-host Shannon? - Does Shannon support bring your own key (BYOK)? - Does Shannon output SARIF? - Which AI providers does Shannon support? - Can I run Shannon on a local or self-hosted model? - Does Shannon actually exploit vulnerabilities, or just scan? What is Shannon? Shannon is an autonomous AI pentester developed by Keygraph. It performs security testing of web applications and their underlying APIs by combining source-code analysis with live exploitation. Shannon analyzes your web application's source code to identify potential attack vectors, then uses browser automation and command-line tools to execute real exploits against the running application and its APIs. Only vulnerabilities with a working proof-of-concept are included in the final report. Shannon is the agent. This repository is Shannon, the open-source pentester you run yourself. The same Shannon also powers Keygraph's commercial editions, Pro and Enterprise. See Editions for how Shannon compares with Pro, Enterprise, and the Community Program. Why Shannon Exists Thanks to tools like Claude Code and Cursor, your team ships code non-stop. But your penetration test? That happens once a year. This creates a massive security gap. For the other 364 days, you could be unknowingly shipping vulnerabilities to production. Shannon closes that gap by providing on-demand, automated penetration testing that can run against every build or release. Why "Shannon"? It's named after Claude Shannon, the father of information theory. At its core, pentesting is an information problem: every probe reduces uncertainty about a system's state. The best tools maximize the signal gained from every request
Top issues
#512[BUG]: Error while running the scan3 comments · 2026-11-03#509[FEATURE]: add config parameter MAX_CONCURRENT_PIPELINES3 comments · 2026-10-31#469Browser-driven phases fail on AppArmor Linux hosts: Chromium sandbox denied in worker (no way to pass apparmor=unconfined / --no-sandbox)2 comments · 2026-09-23#517Report agent never runs set-report-meta with custom models.json provider (DeepSeek) — finalization-report-meta-malformed1 comments · 2026-11-08#479Docker image keygraph/shannon:3.3.0 has 462 critical / 1179 high CVEs (mostly Chromium 149.0.7827.53)1 comments · 2026-10-05#529[BUG]: Exploit stage silently succeeds with an empty collector when the exploitation queue is non-empty — externally-exploitable finding class dropped, report shows zero0 comments · 2026-11-14#527[BUG]: Resuming a terminally-failed workspace dies with an opaque Pipeline failed — persisted error.log reason is never surfaced, and classifyWorkspaceLaunch does not gate on resumability0 comments · 2026-11-13#520[FEATURE]: expose worker activity timeout and concurrency as runtime config (not baked into the image)0 comments · 2026-11-11
Discussions
#508Why the scan is partial? Is there timeout?3 comments · 2026-10-30#502php2shell upgrade verification in Shannon0 comments · 2026-10-26#483Anthropic and OpenAI Cyber Verification Programs1 comments · 2026-10-06#470pre-recon fails but I can't figure out why.2 comments · 2026-09-23#451Unable to run Shannon with Azure OpenAI GPT deployment - Scan fails but tokens are consumed1 comments · 2026-09-08#447Running on diffs0 comments · 2026-09-04#439Shannon 3.0 is here0 comments · 2026-09-02#422Need assistance to configure this SSO / 2FA auth method apps0 comments · 2026-08-12
Latest release
v3.5.1 v3.5.1 2026-11-113.5.1 (2026-11-11) Bug Fixes decode git and subprocess output once to preserve multi-byte characters (518) (804b806) validate workspace names and restrict target URLs to HTTP/HTTPS (515) (4cef63e) worker: link the OpenAI cyber-access error to chatgpt.com/cyber (1880af4)