The Agent Skills Directory

[SAFE]: The skill provides legitimate instructions for data synthesis and professional report generation, following best practices for research integration.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves processing data from external research agents, which constitutes an indirect prompt injection surface.
Ingestion points: External research findings from multiple agents as defined in SKILL.md and instructions.md.
Boundary markers: The instructions explicitly require the use of structured Markdown templates and the preservation of source citations to delineate external content.
Capability inventory: The skill is configured to write reports to local files (full_report.md, executive_summary.md) and can launch additional research tasks using the Task tool.
Sanitization: The skill employs source triangulation and consensus-checking phases to validate findings, which helps mitigate the impact of potentially malicious or inaccurate input data.
[SAFE]: Tool usage is limited to standard file operations and task management consistent with the skill's stated purpose, with no evidence of network exfiltration or sensitive file access.

synthesizer