@companion-ai/feynman@0.5.54
The open source AI research agent.
AI Security Review
scanned 13d ago · by lpm-firewall-aiSource inspection resolves the install hook to a Node version check and traces the flagged runtime behavior to intended service authentication, explicit project setup, and disclosed package telemetry. The scanner signals do not establish malicious behavior.
Decision evidence
public snapshot- The only install hook runs a Node version check and exits when the version is unsupported; it does not perform install-time network or file changes.
- The CLI loads its bundled runtime. AlphaXiv authentication is exposed through the dependency and invoked by the explicit login action; its token request targets the AlphaXiv authentication service.
- The inspected prompt and XML parser dependencies expose their declared library entrypoints; the parser entrypoint exports parsing and validation functions.
- The project initialization behavior is registered as an explicit command. It creates AGENTS.md only when absent and keeps an existing file.
- Telemetry uses the declared PostHog host, describes the collected fields in an interactive first-run notice, supports environment-based opt-out, and sends normalized events and errors. No malicious collection or transfer was identified.
- The package has a preinstall hook, dynamic imports, network activity, and telemetry; inspection traced these to a version check, CLI runtime loading, intended service authentication, and disclosed package telemetry.
Source & flagged code
5 flaggedPackage defines install-time lifecycle scripts.
package.jsonView on unpkgSource gates dangerous network, credential, or execution behavior behind CI, host, platform, time, or geo fingerprint checks.
bin/feynman.jsView on unpkg · L23Findings
Affected versions and remediation
This report applies to @companion-ai/feynman@0.5.54.
See all versions of @companion-ai/feynman for other recorded verdicts.
- Review the evidence and your use of @companion-ai/feynman@0.5.54 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.