@p4code/cli@0.8.0
P4 Code coding agent server and command-line client
AI Security Review
scanned 8d ago · by lpm-firewall-aiInspected source ties the scanner's dynamic-code match to a runtime capability check and its agent-configuration hint to reading Claude config context. The worker and client fetch behavior match session-history, media, and file-preview functions; no concrete attack surface was established.
Decision evidence
public snapshot- package.json describes a coding agent CLI and points its p4c command to dist/bin.mjs.
- The flagged worker location derives Claude config and project-directory values; it does not show an agent configuration write.
- The worker accepts only session-message retrieval or session forking and returns the result as JSON.
- The worker's new Function match is inside a try/catch feature check that returns whether the runtime permits it.
- ChatMarkdown fetches a supplied media URL and turns the response into a browser download.
- ComposerPromptEditor fetches a file preview URL with cancellation and optional range headers.
Source & flagged code
13 flagged · loading sourceSource appears to send environment or credential material to an external endpoint.
dist/claudeHistoryWorker-CMjiofbZ.mjsView on unpkg · L11Source matches reverse-shell style process and socket wiring.
dist/claudeHistoryWorker-CMjiofbZ.mjsView on unpkg · L11Findings
Affected versions and remediation
This report applies to @p4code/cli@0.8.0.
See all versions of @p4code/cli for other recorded verdicts.
- Review the evidence and your use of @p4code/cli@0.8.0 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.