@companion-ai/feynman@0.5.10
Research-first CLI agent built on Pi and alphaXiv
AI Security Review
scanned 26d ago · by lpm-firewall-aiSource inspection shows the install hook performs a Node version check, while the flagged agent configuration write is limited to an explicit init command and preserves an existing AGENTS.md. The token use inspected is directed to the Hugging Face API; no concrete attack surface was established.
Decision evidence
public snapshot- The preinstall hook runs a Node version check, and the CLI entrypoint also checks the Node version before loading the CLI.
- The flagged project configuration command is explicitly invoked and creates AGENTS.md only when it does not already exist.
- The CLI loads a working-directory .env file; no malicious use of its contents was identified in the inspected excerpt.
- The Hugging Face tool sends optional Hugging Face tokens as authorization headers on its API requests.
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@0.5.10.
See all versions of @companion-ai/feynman for other recorded verdicts.
- Review the evidence and your use of @companion-ai/feynman@0.5.10 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.