The Agent Skills Directory

Indirect Prompt Injection Surface: The skill is designed to manage searches across external platforms like chat, email, and cloud storage. This creates a vulnerability surface where malicious instructions embedded in documents or messages could influence the agent's behavior during search or summarization tasks.
Ingestion points: The skill guides the agent to process data from various MCP-connected sources including ~~chat, ~~email, ~~cloud storage, and ~~knowledge base (SKILL.md).
Boundary markers: There are no explicit instructions or delimiters defined within this skill to separate data from instructions or to direct the agent to ignore commands found within the retrieved content.
Capability inventory: The skill manages tool execution priorities for searching, reading, and querying records across enterprise search tools.
Sanitization: The skill does not define specific sanitization or filtering logic for the data ingested from these external sources.

source-management