The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external data (documentation and code examples) via the mcp__context7__query_docs tool, which introduces a theoretical surface for indirect prompt injection from the retrieved content.
Ingestion points: External content is fetched in SKILL.md and SKILL_zh.md through the documentation query tool.
Boundary markers: The skill does not define specific delimiters for the fetched content in these instruction files.
Capability inventory: The skill is limited to document retrieval tools and does not include file system writes, network exfiltration commands, or arbitrary code execution capabilities.
Sanitization: No explicit sanitization of the retrieved documentation strings is provided within the instruction markdown.
[SAFE]: The skill configuration and instructions focus exclusively on official documentation retrieval and do not exhibit any signs of malicious intent, credential theft, or unauthorized persistence.

context7