The Agent Skills Directory

[SAFE]: The skill provides a robust framework for auditing and clarifying goals without using any dangerous system commands or network operations. It focuses on linguistic analysis and task structuring within the chat environment.
[PROMPT_INJECTION]: No patterns associated with bypassing safety guidelines or instruction override were detected. The skill includes specific "User Sovereignty" and "Prohibition" sections that restrict the agent from making unauthorized decisions or fabricating information.
[DATA_EXFILTRATION]: There are no indicators of data harvesting or exfiltration. The skill does not contain any network-facing commands (e.g., curl, wget) or access sensitive file paths.
[REMOTE_CODE_EXECUTION]: The skill does not perform any external downloads or execute remote scripts. All logic is contained within the text-based workflow defined in the markdown file.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided context. While this represents a data ingestion surface, the skill implements strict boundary markers and提问纪律 (Question Discipline) that prevent the agent from being easily manipulated by instructions embedded in the processed data. Given that the skill lacks executable capabilities (file writing, network access), the risk is minimal.

dbs-goal