The Agent Skills Directory

[PROMPT_INJECTION]: The skill instructions and test cases were analyzed for patterns that attempt to override agent behavior, bypass safety filters, or extract system prompts. No such patterns were detected; the language is purely instructional regarding the 'checklist decision' framework.
[DATA_EXFILTRATION]: No evidence of hardcoded credentials, sensitive file path access, or network operations was found in the skill or the test prompts.
[OBFUSCATION]: The content is provided in clear text (Chinese and English). No Base64, zero-width characters, homoglyphs, or other obfuscation techniques were identified.
[REMOTE_CODE_EXECUTION]: The skill does not define any external dependencies, package installations, or remote script execution patterns. It is a text-based cognitive guidance skill.
[INDIRECT_PROMPT_INJECTION]: The skill processes user input related to decision-making. While this provides a surface for potential injection from user data, the skill lacks any capabilities (such as tool usage, file writing, or network access) that could be exploited via such an injection. The risk is negligible.
[DYNAMIC_CONTEXT_INJECTION]: No use of the dynamic context syntax (e.g., !command) was found in the SKILL.md file.
[COMMAND_EXECUTION]: There are no shell commands or subprocess calls defined in the skill instructions or execution steps.

checklist-decision