swiftui-gestures
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEMETADATA_POISONINGEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [METADATA_POISONING]: The skill references fictional software versions such as "iOS 26+" and "Swift 6.3" in
SKILL.md. These future-dated or incorrect version claims could cause the agent to generate hallucinated API guidance or fail to correctly audit current code compatibility. - [EXTERNAL_DOWNLOADS]: The skill instructs the agent to utilize and cite
sosumi.aias a primary documentation source. This domain is not an official Apple resource and is not a whitelisted or well-known service provider. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: User-provided interaction plans and code snippets are ingested through the instructions in
SKILL.mdandreferences/gesture-patterns.md. - Boundary markers: Absent. The skill does not define specific delimiters or instructions to ignore embedded commands within the user's interaction requests.
- Capability inventory: No executable scripts or functions involving subprocess calls, file-write operations, or network requests were detected in the skill package.
- Sanitization: Absent. There are no mechanisms described for filtering or sanitizing external content before it is processed by the agent.
Audit Metadata