test-pact
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external source code provided by the user to generate or refactor tests. While this creates a potential surface for instructions embedded in the codebase to influence the agent, this is a standard risk for development-focused skills. The agent is directed to follow specific reference files (
references/consumer.md,references/provider.md) which helps constrain its behavior. - Ingestion points: The skill analyzes a user-provided codebase.
- Capability inventory: Writing and refactoring test code.
- Sanitization: Not explicitly implemented in instructions.
- [SAFE]: The skill contains active security guidance in
references/consumer.md, explicitly instructing the agent and user not to include sensitive data such as tokens or passwords in test files. - [SAFE]: All network-related configurations in the provided examples, such as the
providerBaseUrl, point to the local development environment (http://localhost:8080), posing no risk of data exfiltration.
Audit Metadata