etch
Pass
Audited by Gen Agent Trust Hub on Jul 26, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute the
convex.jarutility via Bash to perform store operations such as integrity validation, garbage collection, and data migration. - [INDIRECT_PROMPT_INJECTION]: The skill instructions include reading data from external database files, which presents a surface for indirect prompt injection if the processed data contains malicious instructions.
- Ingestion points: Data retrieved via
etch readandetch dumpsubcommands inSKILL.md. - Boundary markers: None provided to distinguish between tool output and instructions.
- Capability inventory: Shell command execution via Bash and file system access.
- Sanitization: No explicit sanitization or filtering of database content is defined.
Audit Metadata