@tekmidian/pai@0.36.1
PAI Knowledge OS — Personal AI Infrastructure with federated memory and project management
AI Security Review
scanned 11d ago · by lpm-firewall-aiThe package has a guarded, first-party agent setup capability that merits a warning under the agent integration policy. Inspected sources do not establish unconsented installation-time mutation, credential exfiltration, or another concrete attack.
Decision evidence
public snapshot- The package has no install lifecycle hook; its only publish lifecycle script runs the build.
- The CLI setup command can write PAI-provided instructions into Claude’s user configuration.
- Updating an existing PAI skill prompts the user and preserves the file if declined.
- The session summary hook processes session input and launches a detached background copy to save a summary.
- The setup writes package-owned agent instructions and skills, with prompts before changing existing files.
- The inspected hook locations read PAI environment configuration; they do not establish an unrelated control-surface write or credential transfer.
Source & flagged code
10 flagged · loading sourcePackage source references child process execution.
dist/context-handover-cache-PtNvj_8D.mjsView on unpkg · L636Findings
Affected versions and remediation
This report applies to @tekmidian/pai@0.36.1.
See all versions of @tekmidian/pai for other recorded verdicts.
- Review the evidence and your use of @tekmidian/pai@0.36.1 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.