@ngockhoale/ukit@3.4.30
Install/update an index-first AI workspace for Claude Code, OpenAI Codex and omp (Oh My Pi).
AI Security Review
scanned 12d ago · by lpm-firewall-aiThe inspected source does not establish malicious behavior or an automatic install attack. The decision client has a concrete conditional credential-forwarding risk because a configured endpoint is not visibly bound to the shared gateway token’s intended service.
Decision evidence
public snapshot- The manifest exposes a CLI and lists no npm lifecycle hook; the package has no automatic install trigger.
- The watchdog and decision-coverage files describe a task hook and a CI coverage guard; the probe, benchmark and performance targets are tooling.
- The install helpers handle project files, protect agent paths, preserve settings and snapshot managed files.
- The decision client obtains an endpoint and API key through separate resolvers, then sends a POST request.
- The endpoint resolver accepts a configured URL, and the key resolver can use the existing shared gateway token; no destination allowlist is evident in the inspected code.
- A configured decision endpoint can receive the shared gateway token when the decision client sends its request. The inspected resolver does not restrict that endpoint to the token’s intended service.
Source & flagged code
6 flagged · loading sourcePackage source references dynamic require/import behavior.
scripts/bench/runGold.mjsView on unpkg · L121Package source references weak cryptographic algorithms.
Findings
Affected versions and remediation
This report applies to @ngockhoale/ukit@3.4.30.
See all versions of @ngockhoale/ukit for other recorded verdicts.
- Review the evidence and your use of @ngockhoale/ukit@3.4.30 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.