bingocode@1.1.205
BingoCode - AI-powered coding assistant CLI built on Claude
AI Security Review
scanned 12d ago · by lpm-firewall-aiSource inspection confirms an automatic postinstall write of a bundled Claude skill, with no guard against overwriting an existing skill. This is a guarded-scope extension lifecycle risk warranting a warning; the inspected source does not establish a broader malicious attack.
Decision evidence
public snapshot- package.json runs scripts/install-skills.cjs from postinstall.
- The installer copies bundled skill files into the user’s Claude skills directory, including over existing files.
- The bundled skill directs Claude’s response style while active.
- The plugin handler is called by the plugin install command; it is separate from the postinstall installer.
- The headless print path installs configured plugins during a CLI run.
- The indicated init and insights locations describe user-facing project guidance and skill usage.
- The plugin manager location loads installed plugins for display.
- The inspected binary and utility locations begin with launch or import setup; those excerpts do not establish additional attack behavior.
- The install hook copies one bundled skill directory; inspected code does not show network transfer or execution of the skill during installation.
- The remaining cited locations do not establish a broader agent configuration write or a separate malicious chain.
Source & flagged code
16 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgPackage contains a possible secret pattern.
src/utils/powershell/parser.tsView on unpkg · L1343Findings
Affected versions and remediation
This report applies to bingocode@1.1.205.
See all versions of bingocode for other recorded verdicts.
- Review the evidence and your use of bingocode@1.1.205 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.