markdown-slides
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data (source Markdown files, outlines) which represents an indirect prompt injection surface. An attacker could potentially place malicious instructions within a document being converted to influence the agent's behavior during the transformation process. However, the instructions are strictly scoped to formatting tasks, and the potential risk is inherent to the skill's primary function.
- Ingestion points: Processes 'Source content' (Markdown files, outlines, or structured content) as defined in the Input Requirements section of SKILL.md.
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for the source content.
- Capability inventory: The skill utilizes tool access for 'Read', 'Write', 'Glob', and 'Bash' (used for file verification tasks like 'ls').
- Sanitization: The skill recommends URL-encoding for image paths (e.g., escaping spaces and parentheses) but does not implement general sanitization of the input text content.
Audit Metadata