testing
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by reading local project source files using tools like
Read,Glob, andGrepacross multiple sub-skills (e.g., characterization-test-generator, tdd-refactor-guard). The instructions do not specify boundary markers or sanitization procedures for the content extracted from these files. Given the skill's capabilities to perform file writes (Write,Edit) and execute build commands viaBash(e.g.,xcodebuild), there is a theoretical surface for indirect prompt injection where malicious instructions embedded in a user's codebase could attempt to manipulate the agent's behavior during test generation or execution. - [EXTERNAL_DOWNLOADS]: The
snapshot-test-setupsub-skill references and recommends the use of theswift-snapshot-testinglibrary hosted on GitHub (github.com/pointfreeco/swift-snapshot-testing). This is a well-known, legitimate, and widely used open-source library in the iOS and macOS development communities.
Audit Metadata