@bongos/core@1.20.27
Cloud Bongos — the AI-first build platform core (GDS + platform surfaces + module system), installed as a versioned dependency (ADR 0108).
AI Security Review
scanned 5d ago · by lpm-firewall-aiSource inspection found first-party Claude configuration materialization during upgrade, which carries residual agent setup risk. The inspected behavior did not establish an active attack, so this warrants a warning rather than a block.
Decision evidence
public snapshot- The manifest exposes a CLI and has no lifecycle scripts; setup is invoked through the package’s upgrade flow.
- The upgrade flow materializes package-provided Claude files into an instance unless materialization is skipped.
- The materializer writes files and can create starter Claude settings when none exist.
- The manifest identifies this as the Bongos platform core; this is first-party agent setup, with no source evidence of an attack.
- The inspected settings path preserves an existing settings file, and the skill-lint hook only reads and reports on a touched skill file.
- The runtime hints resolve to ordinary HTTP metrics middleware and a skill-lint command entrypoint.
- The inspected setup can add package-owned files to an instance’s Claude configuration during upgrade, so its effects reach an agent control surface.
- The inspected source did not establish credential disclosure, unrelated account actions, destructive behavior, or a remote payload.
Source & flagged code
23 flagged · loading sourcePackage contains a critical-looking secret pattern.
tests/public_mirror_export.mjsView on unpkg · L103AWS access key ID in tests/public_mirror_export.mjs
tests/public_mirror_export.mjsView on unpkg · L103Findings
Affected versions and remediation
This report applies to @bongos/core@1.20.27.
See all versions of @bongos/core for other recorded verdicts.
- Review the evidence and your use of @bongos/core@1.20.27 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.