j5ik2o-deep-research-read-me
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of instructional markdown and configuration files aimed at improving documentation quality. No malicious patterns or security risks were detected across any of the analyzed files. \n- [NO_CODE]: There are no scripts (Python, JavaScript, shell, etc.) or binary files provided with this skill. The functionality is delivered purely through AI instructions and reference templates. \n- [PROMPT_INJECTION]: The skill's primary function involves analyzing existing README content, which represents a surface for indirect prompt injection. However, because the skill has no tools for code execution, system access, or network communication, this surface poses no significant security risk. \n
- Ingestion points: The agent reads user-provided project context and existing README files (SKILL.md). \n
- Boundary markers: No specific delimiters are mandated in the instructions for separating input data from instructions. \n
- Capability inventory: No tool use or system-level access is requested or defined in the skill logic or YAML configuration. \n
- Sanitization: Content is processed as natural language without programmatic sanitization, which is acceptable given the absence of executable capabilities.
Audit Metadata