st-full-workflow
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes untrusted data from user work orders and workspace configuration files. * Ingestion points: User work orders and configuration files in the ".ai/strikethroo/config/" directory. * Boundary markers: No explicit delimiters or ignore instructions are used for external content. * Capability inventory: The skill can read and write files, perform Git operations, and execute tasks via external AI harnesses. * Sanitization: No sanitization is performed on ingested content.
- [EXTERNAL_DOWNLOADS]: The "scripts/check-for-updates.cjs" script fetches version information from the GitHub API to check for updates.
- [COMMAND_EXECUTION]: The skill executes local Git commands and AI harness CLI tools. It is instructed to run steps sequentially without user input, which reduces oversight.
Audit Metadata