@feathery/react@2.101.0
React library for Feathery
AI Security Review
scanned 1mo ago · by lpm-firewall-aiSource establishes conditional forwarding of an existing service credential to an unrestricted caller-selected endpoint. This warrants a capability warning, while the inspected evidence does not establish malware intent or justify blocking publication.
Decision evidence
public snapshot- AssistantChat accepts a caller-selected baseUrl and copies the existing feathery_session_token cookie into headers when getJwt is absent.
- The chat transport uses baseUrl and those headers, then sends requests through fetch.
- The link-header wrapper returns existing request options unchanged for destinations outside its backend origins; it does not remove the session header.
- The main CommonJS entrypoint publicly exports AssistantChat.
- The built-in form caller selects the API origin, limiting exposure on that normal usage path.
- No preinstall, install, or postinstall hook or runtime self-dependency appears in package.json.
- No hidden unrelated recipient or malware intent was established.
Source & flagged code
7 flaggedPackage source references dynamic code evaluation.
dist/fthry_AssistantChat.rni-LzLM.jsView on unpkg · L10388Package source references dynamic require/import behavior.
dist/fthry_index.DAzjfWxd.jsView on unpkg · L271Findings
Affected versions and remediation
This report applies to @feathery/react@2.101.0.
See all versions of @feathery/react for other recorded verdicts.
- Review the evidence and your use of @feathery/react@2.101.0 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.