The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves ingesting and analyzing user-provided Markdown files to generate visual summaries and image prompts. This behavior creates a surface for indirect prompt injection where malicious instructions embedded in a document could influence the agent's output.
Ingestion points: The skill processes external Markdown files referenced by the user, as specified in the Workflow and Reading the Markdown sections of SKILL.md.
Boundary markers: The instructions do not define specific delimiters or "ignore" rules to prevent the agent from following instructions found within the analyzed documents.
Capability inventory: The skill is restricted to generating text responses within the chat. Analysis of SKILL.md and accompanying files shows no evidence of file system modifications, network exfiltration, or code execution capabilities.
Sanitization: There is no evidence of content sanitization or validation to filter instructions from the data before it is processed by the language model.
[EXTERNAL_DOWNLOADS]: The skill's main documentation file includes a reference to an external image asset.
Evidence: A hero image is referenced in SKILL.md at https://raw.githubusercontent.com/codebeltnet/agentic/main/skills/markdown-illustrator/assets/hero.png. This resource is hosted on GitHub and is associated with the skill's author.

markdown-illustrator