@leanandmean/coding-agent@0.74.1-scramjet.42
Coding agent CLI with read, bash, edit, write tools and session management
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe package contains expected high-risk primitives for a coding-agent CLI, but no lifecycle hook, self-dependency chain, credential exfiltration, or unconsented control-surface mutation. The Doom example and dynamic extension loading do not establish malicious behavior.
Decision evidence
public snapshot- The manifest has no preinstall, install, or postinstall hook.
- Extensions are imported only when the running agent discovers local, global, or configured extension paths.
- Command execution is an exposed runtime feature for extension and coding-agent operations, with shell disabled by default.
- The tool manager downloads fd or ripgrep only after a running interactive session finds the tool missing.
- The added Doom artifact is a bundled WebAssembly example, not an install-time payload.
- No credential harvesting or unconsented exfiltration was found in the inspected paths.
- No package self-dependency or automatic lifecycle execution was present.
- No code writes to foreign AI-agent configuration during npm installation.
Source & flagged code
9 flaggedFindings
Affected versions and remediation
This report applies to @leanandmean/coding-agent@0.74.1-scramjet.42.
See all versions of @leanandmean/coding-agent for other recorded verdicts.
- Review the evidence and your use of @leanandmean/coding-agent@0.74.1-scramjet.42 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.