Adversarial Testing Python Packages

humanbound

Open-source adversarial testing engine, SDK, and CLI for AI agents. Runs locally or against the Humanbound Platform.

proofagent-harness

Open-source test harness for AI agents. Stress-test production agents with adversarial multi-turn scenarios in CI

gauntlet-evals

pip install gauntlet-evals · v0.1.0. Merge-blocking evaluation gates for generative AI features: YAML suites run against any HTTP endpoint or Python callable, fail the build on a miss, and emit both a diffable JSON pack and a reviewer document cross-referenced to California's published GenAI risk framework. Aligned to, never approved by.

mcp-arcade

GameDay testing for MCP servers. Four named experiments over the JSON-RPC wire, in a Docker sandbox, with scripted and Ollama agent seats. Receipts, tapes and a dataset come out. The oracle is the wire, not the scoreboard.

mcp-stress-test

Stress testing framework for MCP security tools - simulating attacks, temporal degradation, and edge cases

invaris-agentsec

Adversarial security testing for AI agents — prompt injection, tool misuse, memory poisoning, and MCP scanning, with CI-ready regression testing and reports.

humanbound-cli

Open-source adversarial testing engine, SDK, and CLI for AI agents. Runs locally or against the Humanbound Platform.

oubliette-dungeon

Standalone AI Red Team Engine - automated adversarial testing for LLM applications

gauntlex-ai

Adversarial Co-Generation Engine — concurrent Builder+Breaker agents for production-grade AI code security

ares-redteamer

AI Robustness Evaluation System

lifeforge

Autonomous flight simulator for AI agents: co-evolutionary adversarial red-teaming with 3D MAP-Elites quality-diversity. Discovers zero-days in LLM agents before production. MCP-native.

rl-envdoctor

Adversarial QA for LLM-RL environments: find out what reward an empty answer earns. Model-free, zero API cost.

verifierlab

Local-first campaigns for stress-testing graders, reward functions, and policy checkers under optimization pressure.

the-red-council

LLM Adversarial Security Arena — Jailbreak → Detect → Defend → Verify