review-and-iterate
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Executes bash scripts to manage telemetry data and perform static code analysis using the grep utility. These scripts are confined to project and configuration directories.
- [DATA_EXFILTRATION]: Transmits skill usage metadata like platform, duration, and status to a remote service via curl. This is performed only after a user-consent step and excludes sensitive code or project data.
- [PROMPT_INJECTION]: The skill's analysis of user-provided source code creates an indirect prompt injection surface. * Ingestion points: Processes all source files in the current directory and the .superstack/build-context.md file. * Boundary markers: Absent; the instructions do not specify delimiters for the processed files. * Capability inventory: Bash subprocesses, file system reads, and network communication. * Sanitization: None; input data is analyzed directly for review purposes.
Audit Metadata