cybersecdirectory

Enoki Labs makes Enoki, a platform that attacks AI agents to find security and safety weaknesses.

Categories
AI and LLM security, Application security testing
Based in
Ghent, Belgium
Founded
2026
Ownership
Private
Certifications
ISO 27001 Implementing

About Enoki Labs

Enoki Labs makes Enoki, a platform that attacks AI agents to find security and safety weaknesses. It sells to teams that build or run AI agents. The company is based in Gent, Belgium, and was founded in 2026. It is operated by EthicGate BV. The platform connects to an agent's endpoint without code changes, an SDK or installation. It learns what the agent can do and writes attacks for that specific agent. Findings include the exact attack, the response, a severity and mappings to OWASP and MITRE ATLAS. After a fix, Enoki re-runs the attack on every release to check the fix still holds. Runs can start from the Enoki platform, a coding agent over MCP or a CI pipeline over the API. Enoki tests security and safety separately. Security checks whether an attacker can make the agent leak, bypass or misuse what it can access. Safety checks whether the agent produces harmful content or advice it should not give. Enoki is a testing platform, not a runtime guardrail.

Products

Other AI security companies

See all AI security companies

Sources

  1. Wikidata
  2. enokilabs.ai
  3. enokilabs.ai/trust