Self-hosted pentest triage loop
Suggested by runchr · today
Draft — a suggested stack, not yet verified by a member.
Autonomous pentest run whose findings get normalized and triaged into a report.
What they did
PentAGI drives the autonomous penetration test against the target scope, pulling concrete payloads and bypass techniques from PayloadsAllTheThings as it probes. Every confirmed issue is pushed into Faraday, which normalizes and deduplicates the raw tool output into a single vulnerability record set. The agent then reads back Faraday's records to write the final triage report with severity and reproduction steps.
How it went
A deduplicated, severity-ranked vulnerability report backed by Faraday records and reproducible payloads.
Worth knowing
Autonomous scanning can be noisy or hit out-of-scope hosts, so scope must be locked down before the run.
Wiring
Connect PentAGI to the target scope and give it PayloadsAllTheThings as its payload reference skill. Configure Faraday as the destination for findings and hand PentAGI the Faraday API credentials. Run the test, then have the agent query Faraday to assemble the report.
Try it yourself
Run a black-box pentest against staging.example.com (only that host and its subdomains). Use PayloadsAllTheThings for payloads, push every confirmed finding into Faraday, then give me a deduplicated severity-ranked report with repro steps.
Tools in this stack
Run this stack
Install the stack into this project. It writes AGENTS.md and the manifest, merges the MCP servers into .mcp.json, and — with --yes — runs the tool installs.
npx runchr install runchr/self-hosted-pentest-triage-loopcurl -s https://api.runchr.com/api/recipes/runchr/self-hosted-pentest-triage-loop/manifestSystem prompt
You are an autonomous penetration-testing operator. You scope strictly to the authorized targets, verify every finding before reporting it, and record confirmed issues in Faraday.
The manifest is also at api.runchr.com/api/recipes/runchr/self-hosted-pentest-triage-loop/manifest.