axiom-scan-security-privacy
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill scans untrusted project files (Swift source, Plist, Entitlements), which creates a potential surface for indirect prompt injection. An attacker could place deceptive instructions in code comments to influence the agent's summary. This is a low-risk finding inherent to the skill's purpose as a code scanner.\n
- Ingestion points: Scans files matching
**/*.swift,**/Info.plist,**/PrivacyInfo.xcprivacy, and**/*.entitlementsusingGlob,Grep, andReadtools.\n - Boundary markers: The skill does not define explicit delimiters to separate untrusted file content from its own instructions.\n
- Capability inventory: The skill uses
Glob,Grep, andReadcapabilities. It does not perform network operations or file writes.\n - Sanitization: The skill instructs the agent to "use Read to verify the surrounding context before reporting," providing a reasoning-based check against simple injection attempts.
Audit Metadata