swift-expert
Fail
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: CRITICALMETADATA_POISONINGEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [METADATA_POISONING]: The skill includes metadata and external links pointing to a domain flagged as malicious by security scanners. * Evidence: author URL
https://github.com/JeffallaninSKILL.mdfrontmatter. * Evidence: Documentation linkhttps://jeffallan.github.io/claude-skills/skills/language/swift-expert/inSKILL.md. - [EXTERNAL_DOWNLOADS]: The skill references an external documentation resource located on a domain (
jeffallan.github.io) that has been identified as hosting malicious content and is currently on a security blacklist. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to build and test external Swift projects, creating a vulnerability surface for indirect prompt injection if project files are attacker-controlled. * Ingestion points: Project source code and configuration files processed during the Architecture Analysis and Implementation phases in
SKILL.md. * Boundary markers: None present to delimit external code from agent instructions. * Capability inventory: Command execution viaswift buildandswift test. * Sanitization: No sanitization or validation of project contents before execution.
Recommendations
- CRITICAL: 1 file(s) identified as malware by FileRep - DO NOT USE
- AI detected serious security threats
- Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata