The Agent Skills Directory

[SAFE]: The skill facilitates the assembly of analysis results into Markdown reports using internal templates and structured input data. A thorough analysis of SKILL.md and the accompanying scripts (generate_report.py, format_table.py, validate_report.py) shows that the skill operates entirely within the agent's local environment without requesting network access or elevated privileges. The use of the Jinja2 templating engine is safe as it renders from hardcoded templates using locally supplied JSON context. Findings related to data ingestion (such as reading findings.json) are consistent with the skill's primary function and include appropriate constraints, such as structural requirements and string truncation for tables. No prompt injection, obfuscation, or persistence mechanisms were detected.

magic-report-generation