stelow-product-evolutionary-principles
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external inputs such as strategic context and market analysis, which can be a vector for indirect instructions.
- Ingestion points: Strategic context, market analysis, and 'Jobs-to-be-Done' data ingested during the 'context' stage workflow (SKILL.md).
- Boundary markers: Absent; the instructions do not specify delimiters or safety instructions to isolate external data from the skill's logic.
- Capability inventory: The skill references file read/write operations and execution of the
scripts/stelowCLI (SKILL.md). - Sanitization: Absent; no validation or escaping mechanisms are described for the ingested content.
- [COMMAND_EXECUTION]: The skill provides shell logic for environment variable detection (
STELOW_WORKFLOW,STELOW_STATE) and instructs the agent to trigger a hand-off block that results in the execution of the vendor-specificscripts/stelowtool. - [DYNAMIC_EXECUTION]: The skill refers to an external vendor script,
scripts/stelow, which is not bundled with the provided files but is required for the workflow's state management functionality.
Audit Metadata