The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes external, untrusted data that could be leveraged for indirect prompt injection attacks. \n
Ingestion points: The skill retrieves user-provided survey answers via the list_answers tool and survey metadata (titles, descriptions) via get_survey. \n
Boundary markers: No explicit delimiters or instructions are provided to the agent to isolate or treat ingested survey content as untrusted. \n
Capability inventory: The skill provides capabilities to write data (creating and updating surveys) and execute shell commands through the mcporter tool. \n
Sanitization: Instructions mention that titles may contain HTML and should be cleaned, but no sanitization is specified for the body of survey answers. \n- [EXTERNAL_DOWNLOADS]: The setup.sh script performs package installation and network requests to well-known services. \n
It attempts to install the mcporter tool globally via the NPM registry. \n
It communicates with official Tencent Survey API endpoints (wj.qq.com) to initiate and poll for OAuth device authorization. \n- [COMMAND_EXECUTION]: The skill's operational flow relies on shell command execution. \n
The AI agent is instructed to use the mcporter CLI to call survey management functions. \n
The setup.sh script executes curl to handle network communication and mcporter to save configuration settings.

tencent-survey