Filtering secrets on the write path: why agent memory deserves more suspicion at ingest than retrieval
Classic-Being-7960 · reddit · 2026-10-26
The author examines a browser-agent hazard: when a user types "remember my password: ..." into a task box wired straight to long-term memory, the initial design stored it — only for it to surface in model prompts weeks later.
The post covers the small amount of code written to prevent that, and argues the write path into agent memory deserves more suspicion than the read path — secrets should be filtered at ingest, not at retrieval. Implementation details and reasoning included.
More from coding & agent
- Coding Agent Deleted a Production Database During Code Freeze: Anaconda's Governance Fix — anacondainc · 2026-10-27
- Claude Projects now lets you drag threads into the message box or @ mention them for precise context — dfeinition · 2026-10-27
- Jane Street's Yinsky shares tips for making codebases agent-friendly: types, property tests, formal methods — sh_reya · 2026-10-27
- Celesto launches CI runners nearly 90% cheaper than GitHub Actions, with agent sandboxes — aniketmaurya · 2026-10-27
- Runway launches App Studio: build no-code creative apps that generate image, video and audio — runwayml · 2026-10-27
- Supabase founders on rebuilding the backend for a billion AI-built apps — Latent Space · 2026-10-27