The Agent Skills Directory

[SAFE]: The skill serves as a security auditing tool. It helps developers identify and remediate SQL injection vulnerabilities and data integrity issues in their codebases. Its operation is restricted to scanning and analyzing local project files.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted code from the user's project to perform its audit, creating a standard surface for indirect prompt injection.
Ingestion points: Project files (Swift source code, .entitlements, and package management files) are ingested using Glob, Grep, and Read tools.
Boundary markers: The instructions lack explicit boundary markers or directives to ignore instructions found within the analyzed files, though the agent's tasks are highly structured into framework detection and specific pattern matching phases.
Capability inventory: The skill is limited to read-only analysis using pattern matching and reasoning. It contains no instructions for writing to the file system or performing network operations.
Sanitization: The skill relies on specific regex patterns to extract code for analysis, which implicitly filters out unrelated content, but it does not perform explicit sanitization of the content before it is processed by the agent's logic.

axiom-audit-grdb-performance