The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill retrieves content from the codebase which could contain adversarial instructions that influence the agent's subsequent reasoning or output.
Ingestion points: Output from the /sourceatlas:pattern tool, which accesses files within the local repository.
Boundary markers: Absent; the skill does not define specific delimiters or instructions to ignore embedded commands within the retrieved code.
Capability inventory: The skill's capabilities are restricted to information retrieval and pattern synthesis; it does not contain code for network operations, file writing, or command execution.
Sanitization: Absent; there is no evidence of sanitization or validation of the codebase content before it is processed by the agent.

pattern-finder