{"slug": "im-building-cerbere-ag-a-security-evidence-layer-for-ai-agents", "title": "I’m building **Cerbère-AG**, a security evidence layer for AI agents.", "summary": "An anonymous developer is building Cerbère-AG, a security evidence layer for AI agents that observes and controls agent actions across tool calls rather than focusing on inputs like prompt injection or jailbreaks. The project's premise is to evaluate whether an agent is safe to let it act, not just safe to talk to, and the developer is seeking design partners and teams running agents in real or realistic environments to test it and surface failure modes.", "body_md": "[I’m building](//app.cerbereag.site) **Cerbère-AG**, a security evidence layer for AI agents.\n\nMost AI security tools focus on what goes **into** the model: prompt injection, malicious inputs, jailbreaks, etc.\n\nI’m focusing on what happens **after the model decides to act**.\n\nCerbère-AG observes and controls agent actions across tool calls, including:\n\nThe idea is simple:\n\n**Don’t just ask whether an agent is safe to talk to. Ask whether it is safe to let it act.**\n\nI’m looking for developers and teams running AI agents in real or realistic environments to test Cerbère-AG and tell me where it fails.\n\nI’m especially interested in **design partners** who can give real-world feedback on agent workflows, policies, approvals, and failure cases.\n\nIf you build AI agents, security tooling, MCP integrations, or autonomous workflows:\n\n→ Give me your feedback: what would you expect a production-grade agent security layer to catch that Cerbère currently doesn't?\n\nAnd if you find the project useful, a ⭐ on GitHub helps people discover it.\n\nI’m more interested in **breaking it and finding its weaknesses** than in compliments.\n\nIf you have an agent that you think could expose a real failure mode, send it my way.", "url": "https://wpnews.pro/news/im-building-cerbere-ag-a-security-evidence-layer-for-ai-agents", "canonical_source": "https://dev.to/christopher_dikesa/im-building-cerbere-ag-a-security-evidence-layer-for-ai-agents-fgh", "published_at": "2026-09-26 19:27:56+00:00", "updated_at": "2026-09-26 20:01:14.983831+00:00", "lang": "en", "topics": ["ai-agents", "ai-safety", "ai-tools", "agent-protocols"], "entities": ["Cerbère-AG", "GitHub", "MCP"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/im-building-cerbere-ag-a-security-evidence-layer-for-ai-agents", "markdown": "https://wpnews.pro/news/im-building-cerbere-ag-a-security-evidence-layer-for-ai-agents.md", "text": "https://wpnews.pro/news/im-building-cerbere-ag-a-security-evidence-layer-for-ai-agents.txt", "jsonld": "https://wpnews.pro/news/im-building-cerbere-ag-a-security-evidence-layer-for-ai-agents.jsonld"}}