Humanbound
Adversarial testing, a runtime firewall and monitoring for AI agents and chatbots, with an open-source engine
Humanbound tests AI agents and chatbots with adversarial attacks before release, then guards and monitors them in production. Tests run single-turn, multi-turn and agentic attacks (prompt injection, jailbreaks, data exfiltration, tool abuse) and tag findings to the OWASP LLM and Agentic Top 10.
The testing engine, CLI and firewall are open source under Apache-2.0 and run locally with no account. The firewall has four tiers, from input sanitization up to an LLM judge.
The hosted platform adds monitoring and team seats. The Community plan is €0; Team is €349/month.
Pricing: Monthly subscriptions
Choosing a guardrails tool right now?
One click, anonymous. Helps us know our visitors.
Thanks, that helps.
Choosing a guardrails tool right now?
One click, anonymous. Helps us know our visitors.
Thanks, that helps.
Humanbound Alternatives
Explore 61 products in the Observability & Analytics category. View all Humanbound alternatives.
NeMo Guardrails
NVIDIA toolkit for adding programmable guardrails to LLM conversational apps
Guardrails AI
Open-source framework for adding input and output validators around LLM calls
Presidio
Microsoft open-source SDK for detecting and anonymizing PII in text and images
Confident AI
Hosted evals, tracing and monitoring for LLM apps, from the team behind DeepEval
Roder
EU-hosted AI gateway that checks prompts, responses and MCP tool calls against policy, with DLP and team budgets
PIC Standard
Open-source check that blocks an AI agent's tool call unless signed evidence backs it
Work on Humanbound? Feature it at the top of Observability & Analytics.
Is your product missing?