@companion-ai/feynman@0.3.38
Research-first CLI agent built on Pi and alphaXiv
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe preinstall hook only checks Node compatibility. Potentially invasive Pi setup, optional Homebrew invocation, skill installation, and telemetry occur only through explicit runtime/user-command paths and are package-aligned; no concrete malicious chain was found.
Decision evidence
public snapshot- Preinstall only runs the Node-version checker.
- Pi patching is triggered by the user-invoked feynman binary, not npm lifecycle.
- Runtime telemetry targets PostHog and labels Pi capture metadata-only.
- No install-time downloader, credential harvesting, or exfiltration was found.
- The bundled archive is a 59.8 MB runtime workspace with a recorded SHA-256, not an inert hidden payload.
- Runtime writes are package-aligned Feynman/Pi setup; no foreign broad agent-control write occurs during install.
- The skills installer is an explicit user command with a selected target directory.
Source & flagged code
13 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 · L25Findings
Affected versions and remediation
This report applies to @companion-ai/feynman@0.3.38.
See all versions of @companion-ai/feynman for other recorded verdicts.
- Review the evidence and your use of @companion-ai/feynman@0.3.38 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.