knowledge-graph-reindex
Pass
Audited by Gen Agent Trust Hub on Apr 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The script
scripts/reindex.jsutilizeschild_process.execSyncto invoke a local Node.js indexer. The execution path is statically resolved relative to the workspace root, following a standard wrapper pattern. - [DATA_EXFILTRATION]: The skill performs an extensive scan of the workspace, including database schemas, API configurations, and script metadata. While this data is processed locally to generate
knowledge-graph/graph.json, the broad ingestion of sensitive environment data represents a significant metadata exposure surface. - Ingestion points: Workspace-wide scan of skills, scripts, and configuration files as specified in
SKILL.md. - Boundary markers: None identified in the provided wrapper script to delimit untrusted workspace content.
- Capability inventory:
execSynccapability allows for command execution inscripts/reindex.js. - Sanitization: No sanitization or validation of the scanned file contents is performed in the wrapper script before passing control to the indexer.
Audit Metadata