Barrikade Lens

Barrikade Lens is an AI agent discovery and inventory product. It distinguishes autonomous agents, agent-capable tools, model runtimes and supporting development systems. The inventory can include frameworks, MCP servers, skills, models, API services, workflows, repositories and workloads. Discovery sources: - Endpoint collectors inspect known configuration roots, processes and listeners. - Repository and CI scans inspect manifests, imports and validated agent, MCP, A2A, OpenAPI and workflow descriptors. - A Kubernetes controller observes workload metadata with read-only permissions and no Secret or pod-exec access. Evidence and coverage: - Every finding is backed by sanitized evidence. Authoritative descriptors can confirm a system; otherwise independent evidence families are required before confidence is raised. - Possible findings remain labeled as possible rather than being converted into risk scores. - Coverage records show what was checked, what was unavailable and where a collector reported partial results. Privacy: - The data contract accepts inventory identifiers such as hostnames, repository names and sanitized endpoint hosts. - It rejects prompt bodies, source bodies, configuration bodies, credentials, environment values, secret values and full command arguments. Deployment: - Local scans run without signup or network access and do not send telemetry (npx barrikade-lens). - Organization-wide deployments can self-host Lens Hub and export Lens JSON, JSONL or CycloneDX 1.7.