The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple external sources to generate its output, creating a vulnerability surface for instructions embedded in the input data.
Ingestion points: Reads content from transcript.md, research.md, and hooks.md located within the notes-processing/ directory.
Boundary markers: While the skill includes preprocessing steps to strip YAML frontmatter and specific Markdown headers (e.g., ## Related Notes), it lacks explicit delimiters or instructions for the agent to ignore or isolate executable prompts that might be hidden within the raw input text.
Capability inventory: The skill has the ability to read from and write to the local filesystem (specifically notes-processing/{filename}/content.md).
Sanitization: The skill performs basic structural cleaning (converting wiki-links and removing specific sections) but does not validate or sanitize the textual content of the research or transcript files against malicious prompt instructions.

content-writer