The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the wecom-cli tool to perform file operations. It defines specific commands for listing files, searching, downloading, and uploading content to WeCom Disk.\n- [DATA_EXFILTRATION]: The skill facilitates the transfer of files between the local system and WeCom Disk. This behavior is expected for a file management skill and is constrained to the WeCom ecosystem. The instructions explicitly forbid disclosing internal system metadata to users.\n- [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection because it processes untrusted file metadata from an external API.\n
Ingestion points: Data such as file names and search highlights are ingested via the disk files search and disk files list commands in the SKILL.md file.\n
Boundary markers: The agent is instructed to format output in specific markdown structures, but there are no explicit instructions to disregard potential commands embedded in file names.\n
Capability inventory: The skill has capabilities to download, upload, and rename files using wecom-cli across its script definitions in SKILL.md.\n
Sanitization: The skill does not implement specific sanitization logic for data retrieved from the disk service before displaying it to the user.

wecomcli-disk