testing-setup
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and process external data from feature specification files to drive end-to-end testing scenarios. This represents an indirect prompt injection surface where malicious instructions embedded in the specifications could potentially manipulate the agent's behavior during test execution.
- Ingestion points: Acceptance Scenarios (Given / When / Then) located in
docs/features/<feature-id>/spec.md(referenced in SKILL.md Step 3.7). - Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded prompts within the specification files.
- Capability inventory: The agent is instructed to use
@android-clifor executing commands such asandroid run,android layout --diff, andandroid screen capture(referenced in SKILL.md Step 3.7). - Sanitization: There is no mention of sanitization, filtering, or validation logic for the content ingested from the feature specifications.
Audit Metadata