10 open-source GitHub repos for testing prompts, agents, RAG, and red-teaming
victor_explore · x · 2026-11-21
victorexplore (19.5K followers) curated a list of open-source eval and red-team tools on GitHub for AI agent builders (star counts checked Oct 2026):
- Promptfoo (26K★, MIT) — test prompts, agents, and RAG from config; red-teaming/pentesting/vuln scanning; compare GPT/Claude/Gemini/DeepSeek
- DeepEval (19K★, Confident-AI) — unit tests for LLM apps
- SWE-bench (6K★, MIT) — real GitHub issues to test coding agents
- Harbor (6K★, Apache 2.0) — run agent benchmarks in parallel; supports Claude Code, OpenHands, Codex CLI
- Giskard (6K★, Apache 2.0) — scans for wrong answers and risks; red teaming and test generation for agentic systems
- TruLens (4K★, MIT) — score each step with LLM judges
- Inspect (3K★, MIT, UK AISI) — the UK AI Security Institute's evals framework with tool use, multi-turn dialog, and graders
- tau2-bench (2K★, Sierra Research) — agent vs simulated customer for tool-agent-user interaction
- Autoevals (1K★, Braintrust) — ready-made scorers: LLM-as-judge, BLEU/ROUGE, and more
- Scenario (1K★, LangWatch) — simulated users test your agent
Related event: 10 Open-Source GitHub Repos for Testing AI Agents(2 posts)→
More from coding & agent
- Anthropic launches Claude Forward Deployed Engineer path: ~9.5 hours, 11 courses, ends in certification — Redo · 2026-11-21
- LTX 2.5 tested with random prompts via open-source ComfyUI prompt-generation workflow — FeePrestigious7272 · 2026-11-21
- AI first cut generates 5 variants for in-app audition inside Final Cut Pro — Technovangelist · 2026-11-21
- Dev builds an Android port of Hex entirely from his phone, no laptop needed — johnlindquist · 2026-11-21
- 10 open-source GitHub repos to test, eval and red-team your AI agents — victor_explore · 2026-11-21
- Local DeepSeek via Ollama + FCP MCP auto-cuts a 35-min video in ~4 minutes — Technovangelist · 2026-11-21