axiom-audit-grdb-performance
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions define a set of regex-based patterns for identifying performance and security flaws in external Swift code. It does not perform any network operations, exfiltrate data, or attempt to escalate privileges.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external source code, which inherently presents a surface for indirect prompt injection via malicious comments or string literals in the audited files. However, the risk is minimal as the agent is instructed to use specific Grep and Glob patterns for structural analysis rather than interpreting the content as instructions.
- Ingestion points: Reads .swift, .entitlements, and package management files from the project directory.
- Boundary markers: Not explicitly defined for separating code from comments, but reliance on specific regex patterns limits impact.
- Capability inventory: Uses search and read tools (Glob, Grep, Read) for local file analysis only.
- Sanitization: Not applicable as it does not execute the ingested content.
Audit Metadata