The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through tools that read document content. \n
Ingestion points: Tools such as feishu_doc_read and feishu_doc_get_block (referenced in SKILL.md) retrieve untrusted data from Feishu documents into the agent's context. \n
Boundary markers: No specific delimiters or boundary markers are defined in the documentation to distinguish retrieved document content from system instructions. \n
Capability inventory: The skill has significant capabilities to modify the environment via feishu_doc_write, feishu_doc_append, and feishu_doc_delete_block. \n
Sanitization: No sanitization or validation logic for external content is specified in the provided documentation or setup scripts. \n- [EXTERNAL_DOWNLOADS]: Documentation in SKILL.md describes standard dependency installation via pip install -r requirements.txt. \n- [COMMAND_EXECUTION]: The skill includes utility bash scripts (scripts/install.sh and scripts/examples.sh) that perform routine environment checks and print usage instructions; these are standard administrative operations.

feishu-integration