The Agent Skills Directory

[PROMPT_INJECTION]: No malicious override or bypass patterns were detected. The skill uses instructional language like 'CRITICAL CONSTRAINT' and 'IMPORTANT' appropriately to guide the agent's behavior during the interview process, such as limiting questioning to one turn at a time.
[DATA_EXFILTRATION]: No sensitive file access, credential harvesting, or unauthorized network operations were identified. The skill's data handling is restricted to capturing user-provided information during the interview session to build knowledge artifacts.
[REMOTE_CODE_EXECUTION]: The skill does not perform any external downloads, script executions, or package installations. It relies purely on the provided markdown and YAML documentation to function.
[OBFUSCATION]: All analyzed files consist of plain-text markdown, YAML, and XML templates. No Base64 encoding, zero-width characters, homoglyphs, or other obfuscation techniques were detected.
[COMMAND_EXECUTION]: There are no shell commands, subprocess calls, or dynamic code execution patterns. The workflow is restricted to text-based interaction and artifact generation.
[DYNAMIC_CONTEXT_INJECTION]: The skill does not use the dynamic context injection syntax (!command) to execute shell commands at load time.
[SAFE]: The skill represents a well-documented and legitimate tool for research and requirements gathering. The use of XML templates for structured output follows security best practices for data exchange between agents.

research-interviewer