axiom-audit-swiftdata
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user source code, which represents an indirect prompt injection surface. Ingestion points: Phase 1 and 2 in SKILL.md use Glob, Grep, and Read operations on project files. Boundary markers: SKILL.md includes an explicit exclusion list for test files, vendor directories, and build artifacts. Capability inventory: SKILL.md utilizes read-only filesystem tools (Glob, Grep, Read). Sanitization: The skill does not perform explicit content sanitization, relying instead on the agent platform's intrinsic safety controls for tool execution.
- [NO_CODE]: The skill consists entirely of instructional markdown and configuration files without any bundled scripts, binaries, or executable code.
- [SAFE]: No malicious patterns, obfuscation, hardcoded credentials, or unauthorized network activity were identified. The instructions and metadata are consistent with the skill's stated purpose as a SwiftData auditing tool.
Audit Metadata