@p4code/cli@0.8.12
P4 Code coding agent server and command-line client
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe manifest has no install lifecycle hook, and the reviewed worker and agent-configuration locations show expected CLI and provider integration behavior. Scanner flags alone do not establish malicious behavior; the inspected source does not support a concrete attack finding.
Decision evidence
public snapshot- package.json declares a CLI entry point and no lifecycle scripts; the package has no automatic install hook.
- The Claude history worker accepts a session ID and limited history options, then dispatches only session-message retrieval or session forking.
- The worker’s Claude environment lookup reads the Claude configuration directory and project directory name; the inspected code does not establish credential transfer to an unrelated recipient.
- The runtime describes a P4-owned MCP and plugin registry and says provider configuration is read for import.
- The other required files are bundled Effect, schema, client asset, and stdio bridge code; their inspected excerpts do not establish attack behavior.
Source & flagged code
14 flagged · loading sourcePackage contains a possible secret pattern.
dist/runtime-2Dbe0RYi.mjsView on unpkg · L7859Package source references child process execution.
dist/AcpMcpStdioBridge-BL9bAh79.mjsView on unpkg · L4Findings
Affected versions and remediation
This report applies to @p4code/cli@0.8.12.
See all versions of @p4code/cli for other recorded verdicts.
- Review the evidence and your use of @p4code/cli@0.8.12 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.