The Agent Skills Directory

[COMMAND_EXECUTION]: The skill operates by executing shell commands via the feishu CLI tool. It also includes workflow templates in references/workflows.md that utilize standard shell utilities like cat and rm for temporary file management during image processing.- [EXTERNAL_DOWNLOADS]: Installation instructions in references/setup.md direct the user to clone the tool's source code from GitHub and install dependencies via npm. A workflow also suggests the installation of the well-known @mermaid-js/mermaid-cli package for rendering diagrams. These are standard operations for the tool's functionality.- [DATA_EXFILTRATION]: The skill presents an indirect prompt injection surface as it ingests untrusted data from external sources (e.g., feishu doc read, feishu im list, feishu mail list-messages) and possesses the capability to transmit data (e.g., feishu im send, feishu mail send).
Ingestion points: feishu doc read, feishu im list, feishu minutes get, and feishu mail list-messages (documented in references/commands.md).
Boundary markers: Absent. The instructions do not provide specific delimiters to separate agent instructions from the retrieved Feishu content.
Capability inventory: High. The skill can read/write files, perform network API calls, and send messages or emails across all functional areas.
Sanitization: Absent. There is no evidence of content escaping or validation for data retrieved from the Feishu APIs before it is processed by the agent.

feishu-max-saver