The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill converts untrusted external documents (PDF, Word, HTML, etc.) into Markdown for LLM processing. This creates a surface where malicious instructions embedded in a document could influence the AI agent's behavior once the content is converted and read.
Ingestion points: Commands such as uvx markitdown input.pdf ingest content from external files.
Boundary markers: The skill does not provide specific instructions or delimiters to wrap the converted output to prevent the AI from following embedded instructions.
Capability inventory: The skill facilitates reading external files and providing their content to the agent.
Sanitization: No sanitization or filtering of the document content is performed; it is a direct conversion of content to Markdown.
[COMMAND_EXECUTION]: The skill uses uvx to execute the markitdown utility. This is a standard method for running Python tools and interacts with the file system to read input files and write output.

markdown-converter