@p4code/cli@0.8.26
⚠ Under reviewP4 Code coding agent server and command-line client
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe manifest and inspected entry point show a CLI package without install hooks, while the flagged runtime bundle and agent configuration behavior remain unresolved. This review cannot establish a concrete attack, so manual review is warranted.
Decision evidence
public snapshot- The manifest exposes a CLI entry point and declares no install lifecycle script.
- The CLI entry point dispatches to the CLI module; the worker reads the Claude configuration directory.
- The inspected runtime excerpt defines P4 MCP server configuration, but does not establish whether its setup is consent-guarded.
- The inspected excerpts do not establish credential exfiltration or reverse-shell behavior.
Source & flagged code
15 flagged · loading sourcePackage contains a possible secret pattern.
dist/runtime-Cui_uQWU.mjsView on unpkg · L7863Package source references child process execution.
dist/AcpMcpStdioBridge-BuIIHL4p.mjsView on unpkg · L4Findings
Affected versions and remediation
This report applies to @p4code/cli@0.8.26.
See all versions of @p4code/cli for other recorded verdicts.
- Review the evidence and your use of @p4code/cli@0.8.26 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.