pptx-presentation-builder
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's instructions and metadata were analyzed for all 11 threat categories. No prompt injection, obfuscation, credentials, or remote code execution patterns were found.- [INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for untrusted user data to be processed into slides. This surface is managed through structural boundaries. \n
- Ingestion points: User-provided inputs defining the deck brief, audience, and goals (SKILL.md, SKILL.toon).\n
- Boundary markers: The skill uses YAML frontmatter, Markdown code blocks, and clear section headers to delimit user content from core instructions.\n
- Capability inventory: The agent uses standard Read and Write tools to create presentation files and assets as described in the hand-off package section (SKILL.md).\n
- Sanitization: Not explicitly defined, but the skill utilizes rigid templates for slide generation, minimizing the risk of instructions embedded in data influencing the agent's behavior.
Audit Metadata