Strix is an open‑source AI‑driven penetration‑testing platform that runs autonomous agents inside a Docker sandbox to discover, validate, and remediate application vulnerabilities. Users install the tool with a single curl command, then point it at a local directory, a GitHub repository, a live web app, or an API specification using the CLI flag --target. The agents perform reconnaissance, exploitation, and post‑exploitation steps, leveraging a built‑in offensive toolkit that includes an HTTP interception proxy (Caido), automated browser testing, a Python exploit sandbox, and static/dynamic analysis engines. Every finding is accompanied by a working proof‑of‑concept exploit and CVSS‑scored OWASP classification, eliminating false positives typical of static scanners. Strix supports multi‑agent orchestration, allowing specialized AI agents to collaborate in parallel across multiple targets for faster coverage. Results are stored locally under strix_runs/ and can be inspected with the strix view command, which serves a token‑protected dashboard on 127.0.0.1 (by default) or on a specified host/port. For CI/CD, a lightweight GitHub Actions workflow checks out the repository, installs Strix via the same curl script, and runs a non‑interactive quick‑mode scan (strix -n -t ./ --scan-mode quick) on each pull request, blocking insecure code before merge. The tool requires Docker and an LLM API key; supported providers include OpenAI, Anthropic, Google, and OpenRouter, with the default model openrouter/z-ai/glm-5.3. Optional enterprise features add SSO, compliance‑ready reports, VPC/self‑hosted deployment, and BYOK model support. Additionally, the npx skills add usestrix/strix command installs nine agent‑compatible skills for running pentests, applying fixes, and enabling continuous scanning in development environments.