osborn@0.9.249
Voice AI coding assistant - local agent that connects to Osborn frontend
AI Security Review
scanned 22d ago · by lpm-firewall-aiSource shows global agent configuration mutation through explicit CLI execution, which merits a behavior-specific warning under the supplied policy. The inspected behavior does not establish malicious intent or a block-eligible installation attack.
Decision evidence
public snapshot- The CLI launches dist/index.js, which imports dist/claude-llm.js.
- Module initialization changes global Claude settings, setting the compaction window to 1000000 and retention to 3650 days.
- The skill_add frontend command writes supplied content into the user's global Claude skills directory.
- package.json has no preinstall, install, or postinstall hook.
- The configuration changes support the package's coding-assistant functionality.
- No concrete malicious installation chain or credential theft was established.
Source & flagged code
8 flagged · loading sourcePackage source references dynamic require/import behavior.
dist/jsonl-search.jsView on unpkg · L34A single source file combines environment access, network access, and code or shell execution with blocking evidence.
dist/index.jsView on unpkg · L2577Source appears to send environment or credential material to an external endpoint.
dist/index.jsView on unpkg · L2Source executes local commands and sends command output to an external endpoint.
dist/index.jsView on unpkg · L2577Source downloads or fetches remote code and executes it.
dist/index.jsView on unpkg · L2A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
dist/session-access.js#virtual:normalized:round1View on unpkgFindings
Affected versions and remediation
This report applies to osborn@0.9.249.
See all versions of osborn for other recorded verdicts.
- Review the evidence and your use of osborn@0.9.249 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.