Agent OPFOR

Red-team any chat interface. Auto-detects chat widgets, runs adaptive attacks, and generates security reports.

OPFOR is an open-source adversary emulation tool for AI agents. The browser extension is the no-code red-team path — anyone on your team can red-team a deployed chatbot without writing code. WHAT IT DOES: • Auto-detects the chat widget on any web page (custom UIs, Intercom, Zendesk, Drift, Salesforce, etc.) • Types attack prompts into the chat as if you were typing them; watches responses • Judges each response with an LLM and scores it pass / fail • Generates a self-contained HTML report you can download and share EVALUATOR COVERAGE: • OWASP LLM Top 10 • OWASP Agentic AI Top 10 • EU AI Act bias suites WHO IT'S FOR: Product managers, designers, QA, security analysts — anyone who can't or won't open a terminal. Everything runs client-side in your browser. The extension uses your own LLM API keys; nothing is sent to an OPFOR backend. OPEN SOURCE: Source, docs, and issues: https://github.com/KeyValueSoftwareSystems/agent-opfor