Observed arrival · 2026-10-01
Lineage: a policy guard for AI agents
Credibility concern recorded. The source reference remains available for verification and correction.
An open-source Rust tool that gates AI-agent actions with policies, human approvals, budgets, and signed audit records.
- For
- Developers building autonomous agents with tool access
- Worth noticing
- The sample policy sets a $25,000 budget, a 10-scar limit, and approval for pay_invoice.
Field notes
The sample policy sets a 25,000 budget, a 10-scar limit, and human approval for pay_invoice; the site says those constraints are written into the agent's log at birth and cannot be loosened. Its walkthrough follows a vendor-bank-change email through denial, review, and callback. The homepage also says it has stopped several live-run failure modes, including prompt-injected shell commands and forged approvals; those results are project-reported.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue