The Agent Skills Directory

[SAFE]: The skill provides structured instructions for summarizing development activities. It defines clear priorities for information sources and output formats without attempting to bypass safety protocols or execute unauthorized commands.
[DATA_EXPOSURE]: The skill requires access to local git logs and modified file contents. This access is limited to the local environment and is essential for the tool's primary purpose of generating work summaries. No network operations or exfiltration patterns were identified.
[PROMPT_INJECTION]: There are no patterns suggesting direct prompt injection, role-play overrides (DAN), or attempts to extract system prompts. The instructions include safety-conscious constraints such as avoiding the fabrication of facts.

ribao