The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from local markdown files which introduces a potential surface for indirect prompt injection. * Ingestion points: Data is read from ./newsletter/issues/[issue_name]/plan.md and ./substack/notes/ideas.md during the context check phase (Step 0). * Boundary markers: The instructions do not define clear delimiters to separate the ingested data from the skill's operational instructions. * Capability inventory: The skill performs local file reads, invokes external content generation skills (research and write), and appends data back to the filesystem. * Sanitization: There is no indication of content sanitization or validation before the ingested data is used to inform subsequent prompts.

write-note