The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides the agent with 'complete tool freedom' (工具完全自由), specifically authorizing the use of 'Bash' to execute steps in the autonomous task queue. This grants the agent a high degree of autonomy to run arbitrary local commands based on its internal reasoning and task decomposition.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch external data using 'WebSearch' and 'WebFetch' and process it autonomously, which creates a significant injection surface. 1. Ingestion points: Untrusted external data enters the agent's context through search and fetch tools during task execution. 2. Boundary markers: The skill instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing this external content. 3. Capability inventory: The skill utilizes powerful tools including 'Bash', 'Write', 'Edit', and 'Sub-agent' spawning. 4. Sanitization: There are no explicit requirements for sanitizing or validating external inputs before they influence subsequent task steps or tool invocations.

auto-task