osborn@0.9.163
Voice AI coding assistant - local agent that connects to Osborn frontend
AI Security Review
scanned 2mo ago · by lpm-firewall-aiNo malicious install-time chain or undisclosed credential harvesting was confirmed. However, user-invoked deletion of a Claude directory and runtime remote shell installation create concrete dangerous capability that warrants a warning.
Decision evidence
public snapshot- Running the CLI removes the user's .claude directory and replaces it with a workspace symlink when that workspace path exists.
- The running agent can download and execute the code-server installer through a shell pipeline after a start-IDE request.
- The agent can upload selected local artifacts to a frontend URL supplied through environment configuration.
- The manifest has only a prepublish-only script, so installation does not run package code.
- The download-and-execute path is reached only at runtime through the IDE feature, not during installation.
- The OpenAI and Recall requests use configured service credentials for the stated voice and meeting features.
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@0.9.163.
See all versions of osborn for other recorded verdicts.
- Review the evidence and your use of osborn@0.9.163 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.