The Agent Skills Directory

[SAFE]: The skill is a technical guide for optimizing the cognee RAG framework and does not contain any executable malicious instructions or injection patterns.
[INDIRECT_PROMPT_INJECTION]: The documentation identifies data ingestion points where the agent processes external files. 1. Ingestion points: Local document directories passed to cognee.remember() (SKILL.md). 2. Boundary markers: The guide focuses on performance and does not specify data isolation markers. 3. Capability inventory: LLM-based extraction, summarization, and vector database storage. 4. Sanitization: Not discussed in this performance-oriented context.
[EXTERNAL_DOWNLOADS]: The skill references standard installation of vendor-related dependencies, such as the gliner extractor for cognee, which can be installed via pip or managed via environment variables.

cognee-performance