mtech-ai-local-runner@0.1.85
D6 owner revision `fbc30fbe-cac9-45d1-9eff-bccec6ee6a54` selects a separate Gateway-initiated P2 verification budget: at most 1,500 weighted minutes per UTC calendar month and one active verification per workspace. This does not cap or certify the entire
AI Security Review
scanned 4d ago · by lpm-firewall-aiThe manifest has no install lifecycle hook, and agent configuration changes are tied to explicit setup or integration commands. The changed Work Room code periodically reads activity and inbox counts from the configured Gateway, consistent with its described feature; no review-directed override or attack behavior was found.
Decision evidence
public snapshot- The package exposes a CLI and has build and command scripts, but no install lifecycle hook.
- The CLI installs agent integrations only through explicit setup or integration commands; the README describes the Work Room pane as part of setup.
- The Work Room session hook refreshes counts from the Gateway, and its code says it reads activity and inbox items, not message bodies.
- The configured Gateway endpoint belongs to the package's stated AI Gateway service.
- After the Work Room plugin is installed, it refreshes activity and inbox counts at session start and periodically.
- Setup can write the package's own Claude plugin; requesting Claude or Codex integrations also writes the runner's hooks to those apps' configuration directories.
Source & flagged code
5 flaggedPackage text addresses the security reviewer or scanner and tries to influence the review outcome.
README.mdView on unpkgPackage source references dynamic require/import behavior.
dist/src/adapters/codex/pre-model-gate.jsView on unpkg · L31Findings
Affected versions and remediation
This report applies to mtech-ai-local-runner@0.1.85.
See all versions of mtech-ai-local-runner for other recorded verdicts.
- Review the evidence and your use of mtech-ai-local-runner@0.1.85 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.