superclawd@0.1.108
Claude Code, supercharged — your team's standards, every session. The superclawd CLI loads your workspace's skills, agents, commands, workflows, and live directives into Claude Code as a plugin.
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe reporting is related to the advertised repository-memory feature, and the source indicates the diff is assessed locally rather than posted directly. However, the package silently installs a persistent commit hook that sends repository-identifying metadata and diff-derived reasons, a runtime reporting behavior not disclosed in the README; this warrants a warning.
Decision evidence
public snapshot- A normal CLI launch in a repository installs a post-commit hook that runs the helper’s report-commit command.
- The hook’s report-commit handler reads the latest commit diff and compares it with saved repository memories.
- The helper posts the commit hash, repository fingerprint, stale memory IDs, and reasons to the SuperClawd MCP API.
- The README describes AI-captured repository memories, but does not mention the automatic post-commit reporting hook.
- The commit check serves the package’s repository-memory feature; source passes the diff to a local Claude CLI for the staleness judgment, while the posted event contains staleness metadata rather than the diff itself.
- The README identifies repository memories as AI-captured and scoped to a git-remote fingerprint.
Source & flagged code
13 flaggedSource sends credentials or rich application records to a package-controlled external receiver enabled by default.
dist/index.jsView on unpkg · L2Findings
Affected versions and remediation
This report applies to superclawd@0.1.108.
See all versions of superclawd for other recorded verdicts.
- Review the evidence and your use of superclawd@0.1.108 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.