The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes potentially untrusted data from task descriptions, calendar events, and contact information retrieved via lark-cli. Instructions embedded within these external fields could potentially influence the agent's output during the analysis phase.
Ingestion points: Data fetched in SKILL.md using lark-cli task +search, lark-cli calendar freebusy list, and lark-cli contact +search.
Boundary markers: The skill does not define specific delimiters or instructions to the agent to disregard commands embedded within the fetched data.
Capability inventory: The skill's capabilities are limited to reading organizational data and generating a markdown report; it does not include tools for writing to the file system or executing arbitrary network requests.
Sanitization: There are no explicit instructions for the agent to sanitize or filter the content of task or calendar entries before processing them.
[EXTERNAL_DOWNLOADS]: The skill references documentation and shared security guidelines from the larksuite organization's public GitHub repository.
Evidence: References to https://github.com/larksuite/cli/blob/main/skills/lark-shared/SKILL.md and related task/calendar skill definitions.
[COMMAND_EXECUTION]: The skill relies on the lark-cli command-line tool to perform authenticated operations and data retrieval.
Evidence: Execution of commands such as lark-cli contact users find_by_department and lark-cli task +search --assignee "<user_open_id>" to gather workload metrics.

lark-workflow-workload-balancer