@tekmidian/pai@1.0.10
⚠ Under reviewPAI Knowledge OS — Personal AI Infrastructure with federated memory and project management
AI Security Review
scanned 2h ago · by lpm-firewall-aiThe inspected excerpts do not substantiate the scanner’s malicious verdict, and several indicators have plausible product-aligned explanations. The large CLI bundle and agent integration paths were not fully traced, so the remaining uncertainty warrants manual review.
Decision evidence
public snapshot- package.json exposes CLI and daemon entrypoints and has a prepublishOnly build command, but no preinstall, install, or postinstall hook.
- The inspected Todoist setup reads a Todoist token and uses the Todoist API; the inspected provider configuration lists Kimi as a configured provider.
- The generated rule source defines a blocking base64-to-shell detection rule, which is consistent with defensive detection rather than proof of a reverse shell.
- The CLI bundle is large, and the credential-redirect finding was not fully resolved across all login and recovery callers.
- Agent configuration targets and the remaining flagged runtime locations were not traced through all callers and activation paths.
Source & flagged code
17 flagged · loading sourcePackage source references child process execution.
dist/project-setup-MA5XUrs3.mjsView on unpkg · L28Package source references weak cryptographic algorithms.
dist/project-setup-MA5XUrs3.mjsView on unpkg · L28Findings
Affected versions and remediation
This report applies to @tekmidian/pai@1.0.10.
See all versions of @tekmidian/pai for other recorded verdicts.
- Review the evidence and your use of @tekmidian/pai@1.0.10 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.