test
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources to define its execution plan, creating a surface for instructions embedded in data to influence agent behavior.
- Ingestion points: Reads task definitions from
docs/task/{ID}-{task-name}.md,AGENTS.md, and tracking information fromTASKS.md. - Boundary markers: The instructions do not define delimiters or warnings to ignore instructions found within the processed files.
- Capability inventory: The skill executes shell commands for unit, integration, browser, API, and CLI checks, and performs file write operations to
docs/testing/andTASKS.md. - Sanitization: There is no mention of filtering or validating the verification approach or acceptance criteria extracted from task documents before they are used to run commands.
- [COMMAND_EXECUTION]: The skill is designed to run arbitrary project-specific verification commands. While intended for testing, this capability could be misused if the instructions in the task documents are manipulated to execute malicious payloads during the 'verification' phase.
Audit Metadata