The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local search and file inspection utilities including find, rg (ripgrep), head, and cat to locate and read database schemas, Drizzle ORM models, and migration scripts. These commands are statically defined and scoped to the project's service directories.
[PROMPT_INJECTION]: The skill exposes an indirect prompt injection attack surface as it reads and summarizes content from codebase files. 1. Ingestion points: Repository files, SQL migrations, and Drizzle schema files identified via shell commands. 2. Boundary markers: The skill instructions do not define specific delimiters or instructions to ignore embedded commands within analyzed files. 3. Capability inventory: The agent utilizes local filesystem discovery and file reading capabilities to process project data. 4. Sanitization: No specific filtering or sanitization is applied to the ingested code or SQL content before generating deliverables.

persistence-analyst