The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The pattern describes a system that ingests external codebase data, which creates a potential surface for indirect prompt injection. * Ingestion points: Codebase files accessed via the retrieveFiles function described in SKILL.md. * Boundary markers: None identified in the provided prompt integration guidelines. * Capability inventory: File retrieval and relevance scoring logic. * Sanitization: No specific content sanitization or filtering logic is mentioned in the pattern.
[SAFE]: The skill consists entirely of conceptual documentation and illustrative code snippets. There are no executable files, remote downloads, or signs of malicious intent or obfuscation.

iterative-retrieval