The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the wecom-cli binary for all document management tasks and instructs the agent to use curl for downloading images from WeCom CDN URLs during content analysis.- [EXTERNAL_DOWNLOADS]: As part of its document parsing workflow, the skill downloads image assets from WeCom's content delivery network (e.g., w...qpic.cn) to the local environment.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process content from external Smartpage documents, which are untrusted data sources.
Ingestion points: Document content is retrieved using the smartpage pages get command as described in references/smartpage-edit.md.
Boundary markers: While the skill uses structural MDX tags for formatting, it lacks explicit safety delimiters to isolate ingested text from the agent's instructions.
Capability inventory: The skill possesses capabilities for writing to the file system, modifying remote documents via wecom-cli, and performing network operations via curl.
Sanitization: The skill includes detailed safety guidelines in SKILL.md that explicitly require the agent to block XSS patterns (like <script> or javascript:), political sensitivities, and role-play injection attempts.- [DYNAMIC_EXECUTION]: The skill utilizes a domain-specific formula system (documented in references/formula-reference.md) that evaluates logic and performs operations like ADDRECORD and MODIFYRECORDS at runtime based on page控件 (controls) and document data.

wecomcli-smartpage