The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from Feishu Bitable records, which creates a surface for potential indirect prompt injection.
Ingestion points: The feishu_bitable_app_table_record.list tool (referenced in SKILL.md) allows the agent to pull external record data into its context.
Boundary markers: The skill instructions do not prescribe specific delimiters or markers to isolate ingested record content from the agent's system instructions.
Capability inventory: The skill possesses broad capabilities, including creating, updating, and deleting records and tables, as well as uploading files via feishu_drive_media.
Sanitization: The documentation focuses on type-level validation (e.g., requiring millisecond timestamps and specific JSON structures) rather than content sanitization to prevent embedded instructions from influencing agent behavior.

feishu-bitable