The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content from external Feishu documents and wikis, which is then integrated into the agent's context. This presents a surface for indirect prompt injection if the source documents contain malicious instructions.
Ingestion points: Documents are retrieved using feishu-cli doc export and feishu-cli wiki export (read/workflow.md).
Boundary markers: There are no instructions for the agent to use specific delimiters or to ignore potential instructions embedded within the retrieved data.
Capability inventory: The agent can execute shell commands via feishu-cli, perform file writes, and execute Python code for validation.
Sanitization: A Python script is provided to validate the UTF-8 encoding of Markdown files to prevent processing malformed data (write/workflow.md).
[COMMAND_EXECUTION]: The skill relies on the execution of the feishu-cli binary and other shell utilities like jq and python3 to perform its core functions. These tools have permissions to interact with the file system and network.
[DYNAMIC_EXECUTION]: The workflow includes a Python one-liner to check file encoding (python3 -c ...). While intended for validation, this constitutes the dynamic execution of code strings.
[EXTERNAL_DOWNLOADS]: The skill references an external GitHub repository (github.com/riba2534/feishu-cli) for the installation of the core feishu-cli tool. The repository belongs to the skill's author.

feishu-cli-docs