@aspectlylabs/mikka@0.0.332
Local Mikka daemon: runs agents through the user's own CLIs
AI Security Review
scanned 2d ago · by lpm-firewall-aiThe source behavior matches a local agent daemon that runs the user's CLIs and routes model requests to the selected provider. The flagged subprocess and network behavior serves that purpose, and the reviewed persistence path withholds model credentials.
Decision evidence
public snapshot- The manifest describes a local daemon that runs agents through the user's own CLIs; the README describes connecting the computer to Mikka and running Claude Code or Codex.
- The model broker defaults to Anthropic's or OpenAI's API and accepts a provider URL only over HTTPS.
- The broker adds the key to the provider authorization header and disables redirects.
- Run journal persistence removes the model credential from the saved assignment.
- CLI execution and model API access match the documented daemon purpose.
- The reviewed credential flow routes keys to the selected provider endpoint; the run journal omits the key.
Source & flagged code
6 flaggedPackage source references child process execution.
github-credentials.mjsView on unpkg · L7Findings
Affected versions and remediation
This report applies to @aspectlylabs/mikka@0.0.332.
See all versions of @aspectlylabs/mikka for other recorded verdicts.
- Review the evidence and your use of @aspectlylabs/mikka@0.0.332 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.