openauditB85out of 100: openaudit's audit of itself
agent.deepsec

deepsec AI vulnerability scan

Vercel Labs deepsec: regex candidate scan followed by an AI agent review of the riskiest files (or of changed files with settings diff). Opt-in: run with --only agent.deepsec. Costs real money (default cap about $5, enforced between chunks) and the review agent gets a sandboxed shell in the target, so only run it on code you trust. State goes to a temp directory, not the target.

Engine
Agent
Wraps
deepsec
Targets
Codebase
Areas
Security
Runs
Opt-in: only when selected with --only
Network
Contacts services beyond the target

Run only this auditor

bunx openaudit . --only agent.deepsec

Runs a playbook with the Claude Agent SDK. Needs ANTHROPIC_API_KEY or a Claude Code login; host agents can run the same playbook over MCP without a key.