The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to process external research findings, which inherently creates a surface for indirect prompt injection. However, the instructions provide a structured framework for data validation, including source quality assessment and consensus building (documented in Phase 1 and 2 of instructions.md). The ingestion points are the research findings, and while explicit boundary markers are absent, the skill's capabilities are limited to writing standard report files and launching specific research tasks. Sanitization is addressed through the mandatory quality and credibility evaluation phase.
[SAFE]: Analysis of the skill's logic and instructions confirms that it does not attempt to exfiltrate data, execute unauthorized commands, or bypass safety guidelines. The tool usage is consistent with its stated purpose of research synthesis.

synthesizer