The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process various external 'authoritative artifacts,' including prior drafts, protocols, and structured result summaries.
Ingestion points: Manuscript briefs, prior drafts, protocols, and research notes are identified as inputs in SKILL.md and references/workflow.md.
Boundary markers: While the skill instructs the agent to record missing evidence, it lacks specific technical delimiters to isolate content from the processed artifacts.
Capability inventory: The skill has the capability to call external lookup tools (e.g., /crossref-lookup) and generate comprehensive manuscript files.
Sanitization: The instructions do not define methods for sanitizing or validating the content of artifacts before they are integrated into the writing process.

scientific-writing