Kastra: Policy Interception for AI Coding Assistants
carlosjimenez1 · hn · 2026-07-09
The team launched Kastra, a runtime authorization and policy enforcement tool built for AI coding assistants. It intercepts Agent tool calls, making deterministic allow, pause, or deny decisions before dangerous actions (like accidentally deleting a production database) occur.
- Pain Point: LLMs are probabilistic, meaning prompts alone cannot fully constrain Agent behavior. The team once faced a near-miss where Cursor almost deleted a production database, realizing their stack lacked deterministic controls for Agent permissions.
- Core Features: Offers a web interface to write policies in natural language, includes built-in policy packages for high-risk scenarios, and records immutable audit logs.
- Recon Scan: Scans local Agent history to automatically identify past dangerous actions and generate protective rules. Its desktop and CLI features are free for developers.
More from coding & agent
- Kimi K3 rises to No. 4 on the Agent Arena leaderboard — HeyZoyaKhan · 2026-07-22
- Claude adds screen-recorded skills that can replay your workflow — CodeByPoonam · 2026-07-22
- Devin adds e2b sandboxes for remote agent execution — badphilosopher · 2026-07-22
- Hermes Agent Refactoring Proposal: Decoupling via Event Bus and Monorepo Slicing — Promptmethus · 2026-07-22
- ty now reads Pydantic config keywords and field metadata — charliermarsh · 2026-07-22
- Pensar Launches AI Security Agent to Autonomously Discover and Patch 0-Days — andriy_mulyar · 2026-07-22