Mnestiq releases open-source tamper-evident recorder for AI agent actions
Start here
Mnestiq released v1.1.1, an Apache-2.0 open-source Python tool that records what AI agents saw, decided and did as signed, tamper-evident evidence. Records are hash-chained and sealed with Ed25519-signed Merkle checkpoints, so edits, deletions or reordering fail verification. The tool also maps a firewall or flow-log connection back to the exact agent run and step that opened it, and supports Anthropic, OpenAI, Azure OpenAI and compatible APIs. The evidence format, recorder, verifier and dashboard are free and open source. According to Mnestiq, the dashboard runs only on the user's machine, and prompts and data stay on the user's own systems, with only signed checkpoint fingerprints leaving and only to destinations the user configures. Mnestiq says it is seeking a small number of security and platform teams to shape the workbench around real incidents.
Generated from reports · updated 23 min ago
Report timeline
Follow the reports to see the event from different sides.
- Hacker News · AIShow HN: Mnestiq links firewall log entries to the AI agent step behind them
Mnestiq, an open-source Python tool released as v1.1.1 under Apache-2.0, records what AI agents saw, decided and did as signed, tamper-evident evidence, then maps a firewall or flow-log connection back to the exact agent run and step that opened it. Records are hash-chained and sealed with Ed25519-signed Merkle checkpoints, so edits, deletions or reordering fail verification, and it supports Anthropic, OpenAI, Azure OpenAI and compatible APIs.
Heat for this event
Not enough continuous observations yet to draw a trend.