eve@0.82.0
Filesystem-first framework for durable backend AI agents that run anywhere.
AI Security Review
scanned 1d ago · by lpm-firewall-aiSource inspection found no concrete attack behavior. The scanner’s credential-transfer, dynamic execution, and similarity hints resolve to Vercel-scoped authentication, CLI bootstrap behavior, and ordinary framework or protocol code.
Decision evidence
public snapshot- package.json lists build and test scripts, with no install lifecycle hook; bin/eve.js runs the CLI after direct invocation.
- Vercel OAuth endpoints are restricted to Vercel hosts, and the access token is sent to Vercel’s AI Gateway credits endpoint.
- The flagged UI chunks subscribe to and update local agent stores.
- The HTTP basic module compares supplied credentials with its configured strategy; the skills module validates resource paths.
- The flagged dependency files contain HTTP, WebSocket, encoding, parsing, and framework runtime code.
- The flagged network policy file only exports a baseline policy constant.
Source & flagged code
8 flagged · loading sourceSource spawns a local helper that also contains network and dynamic execution context; review data flow before blocking.
bin/eve.jsView on unpkg · L2Package source references dynamic require/import behavior.
bin/eve.jsFindings
Affected versions and remediation
This report applies to eve@0.82.0.
See all versions of eve for other recorded verdicts.
- Review the evidence and your use of eve@0.82.0 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.