Perplexity Open-Sources Numbat: Endpoint Visibility Tool for AI Agent Activity
eschadiol · x · 2026-07-31
Perplexity has open-sourced a new tool called Numbat on GitHub, designed to provide visibility into AI agent activities on local endpoints. The tool features on-device detection, optional pre-action blocking, and forensic reconstruction capabilities.
Numbat observes supported desktop, CLI, IDE, and gateway agents through local hooks, plugins, and OTLP/HTTP logs. It normalizes both live and at-rest activities into a single event model and evaluates them using a CEL rule engine, ensuring secure and controllable operations for AI agents executing tasks on endpoints.
Related event: Perplexity Open-Sources Numbat for Endpoint AI Agent Security(6 posts)→
More from coding & agent
- FDE-Gym benchmark: best agent fully solves only 14 of 95 real customer-delivery tasks — dair_ai · 2026-10-16
- Deep Research API showdown: OpenAI, Perplexity, Parallel, Tavily, Gemini, Valyu compared — bar_raiser333 · 2026-10-16
- Geoffrey Huntley open-sources Overalls, a self-hosted AI developer platform under MIT — kieranklaassen · 2026-10-16
- Dev shows 199.7bn tokens over 200 days on OpenAI Codex, peak 8.6bn/day — Dimillian · 2026-10-16
- Ask Claude to open its changes in nvim and scroll to the exact line for review — max__drake · 2026-10-16
- Proof CEO on x401: KYC-compatible agent identity using NIST IAL2 verifiable credentials, composing with x402 payments — csuwildcat · 2026-10-16