The Agent Skills Directory

[PROMPT_INJECTION]: Instructions in the SKILL.md file specifically target the AI assistant to override standard authorization flows. These instructions include directives like 'I have authorized' and 'do not confirm for me,' which are designed to suppress interactive safety prompts during sensitive operations.\n- [REMOTE_CODE_EXECUTION]: The documentation describes a procedure to fetch and execute packages from the official Alipay NPM registry using npx -y @alipay/agent-payment@latest install. This pattern involves the execution of remote code within the agent environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill's input processing mechanism in scripts/engine/store-sales-report-check.js lacks sanitization. It ingests arbitrary text from the user or files, which could contain malicious instructions. If the analysis results are subsequently processed by an LLM, these instructions could lead to indirect prompt injection.

store-sales-report-check-free