osborn@0.9.260
⚠ Under reviewVoice AI coding assistant - local agent that connects to Osborn frontend
AI Security Review
scanned 2d ago · by lpm-firewall-aiThe inspected source does not establish malicious behavior, and the manifest has no automatic install hook. Required runtime dependencies remain unresolved or only partly inspected, so this review cannot support a clean verdict.
Decision evidence
public snapshot- package.json defines prepublishOnly as a build command and declares the Anthropic agent SDK and LiveKit dependencies.
- dist/claude-llm.js imports the agent SDK, and the available SDK manifest identifies sdk.mjs as its runtime entry point; the full dependency runtime chain remains unresolved.
- dist/index.js runs the code-server installer through a shell when code-server is missing; this is a sensitive installation behavior, but the inspected source does not establish an attack.
- The inspected manifest has no preinstall, install, or postinstall hook.
- The inspected upload code sends artifacts to a frontend-configured upload endpoint; no unrelated credential recipient was established.
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@0.9.260.
See all versions of osborn for other recorded verdicts.
- Review the evidence and your use of osborn@0.9.260 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.