The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides shell script templates in references/delegation-context-template.md, references/editorial-pipeline.md, and references/style-guide.md for verifying file structures and enforcing style rules using find and grep. These commands operate on user-influenced variables like <draft-slug> and <path>, which could be exploited for command injection if the agent fails to apply the recommended sanitization (e.g., kebab-case conversion).
[INDIRECT_PROMPT_INJECTION]: The editorial pipeline processes external draft content, creating a vulnerability surface where malicious instructions within a draft could attempt to override editorial guidelines or manipulate the agent's reporting behavior.
Ingestion points: Draft file content and absolute paths ingested during the editorial passes defined in references/delegation-context-template.md and references/editorial-pipeline.md.
Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings for the draft content being processed.
Capability inventory: The skill allows the agent to read and modify local files (during the humanize pass) and execute shell utilities (find, grep, wc) for verification.
Sanitization: No content sanitization or escaping mechanisms are described for handling potentially malicious text within the drafts.

editor-methodology