scan-all
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from project reference documents and custom files during the refresh workflow.
- Ingestion points: Content is read from
referenceDocsand user-defined documentation files. - Boundary markers: The protocol does not define explicit delimiters or instructions to ignore embedded commands within the processed files.
- Capability inventory: The skill is authorized to use
spawn_agentand execute shell commands vianodefor lifecycle hooks. - Sanitization: No specific input sanitization or filtering logic is described for the document contents being scanned.
- [COMMAND_EXECUTION]: The skill executes local scripts via the shell to manage project state, specifically invoking
node -eto call logic within./.claude/hooks/lib/session-init-helpers.cjs.
Audit Metadata