osborn@0.9.169
Voice AI coding assistant - local agent that connects to Osborn frontend
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThis is a concrete remotely reachable data-exposure vulnerability in a user-invoked service, not evidence of an automatic malicious install chain. The absence of install hooks prevents a publish block, but the exposed session archives warrant a warning.
Decision evidence
public snapshot- The runtime server permits every web origin and defaults to binding on all interfaces.
- An unauthenticated session-list endpoint discloses session IDs, paths, and last messages.
- The export-one endpoint uses those IDs to archive Claude session files without authentication.
- The full export is also unauthenticated unless an optional environment token is configured.
- The manifest has only a prepublish hook, not an install-time lifecycle hook.
- The main program starts only through the user-invoked CLI.
- The observed OpenAI request is a TTS request using the configured API key, not credential harvesting.
Source & flagged code
8 flagged · loading sourcePackage source references dynamic require/import behavior.
dist/jsonl-search.jsView on unpkg · L34Findings
Affected versions and remediation
This report applies to osborn@0.9.169.
See all versions of osborn for other recorded verdicts.
- Review the evidence and your use of osborn@0.9.169 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.