mtech-ai-local-runner@0.1.80
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 5d ago · by lpm-firewall-aiThe package has no install lifecycle hook, and its CLI starts only on direct invocation. The reviewed Codex flow validates a loopback MCP endpoint and a restricted local session before a model turn; the scanner’s review-related README match appears to describe product behavior rather than direct reviewer manipulation.
Decision evidence
public snapshot- The manifest has no install, preinstall, or postinstall hook; its scripts build or run explicit CLI commands.
- The CLI calls main only when invoked directly, so importing it does not start the runner.
- The Codex pre-model gate accepts only a loopback HTTP endpoint at 127.0.0.1 with the expected /mcp path.
- Codex sessions require an ephemeral, read-only, network-disabled local job profile and validate policy and tool inventory before sending a model turn.
Source & flagged code
6 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@0.1.80.
See all versions of mtech-ai-local-runner for other recorded verdicts.
- Review the evidence and your use of mtech-ai-local-runner@0.1.80 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.