osborn@0.9.175
Voice AI coding assistant - local agent that connects to Osborn frontend
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe high-risk primitives are reachable through the user-invoked assistant rather than an install hook, and the inspected OpenAI call sends speech input rather than harvested environment data or command output. No automatic lifecycle chain, remote payload execution, or concrete exfiltration behavior was found.
Decision evidence
public snapshot- The manifest exposes a CLI and has only a prepublish build hook, not an install lifecycle hook.
- The CLI starts the packaged application only after the user invokes the osborn command.
- The runtime sends supplied speech text to OpenAI's speech endpoint using the configured API key.
- The session helper reads Claude session data from the configured or default Claude directory.
- The Claude integration changes two compaction settings when the user-run application starts.
- No preinstall, install, or postinstall script is present.
- No self-dependency or download-and-execute path was found.
- The network use inspected is a direct, package-aligned text-to-speech request rather than credential or command-output exfiltration.
- The shell approval shown is restricted to localhost canvas and TTS routes without shell chaining.
Source & flagged code
9 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@0.9.175.
See all versions of osborn for other recorded verdicts.
- Review the evidence and your use of osborn@0.9.175 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.