# Pensar Apex

A terminal-based, AI-agent-driven penetration testing tool that runs blackbox or whitebox tests against a target, and reports vulnerabilities with severity, evidence and suggested fixes.

Pensar Apex runs penetration tests from the terminal using autonomous agents that explore a target, reason about what they find, and surface real vulnerabilities, in either blackbox or whitebox mode using the target's source code.

A slash command runs a pentest before merging a pull request, returning findings with severity scores, evidence and suggested fixes without requiring security expertise, and the same tests can run headless in CI/CD. An operator mode supports manual investigation, exploit chaining and validation, with persistent memory that accumulates across engagements, and testing can scale across an attack surface with agent-driven swarm testing.

It is a TypeScript command-line tool with a terminal UI and a headless mode with flags for target, mode, model and custom guidance, and can use a custom OpenAI-compatible inference endpoint in place of its default model.

Website: https://pensar.dev