The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to implement its diff-check mode. This mode executes git commands to retrieve changes from specific commits or ranges, allowing the agent to perform linting on newly added text lines. This is a functional use of shell tools for a development-oriented writing workflow.
[INDIRECT_PROMPT_INJECTION]: As a text processing tool, this skill possesses an inherent attack surface for indirect prompt injection by reading untrusted content from user-provided files.
Ingestion points: Text is ingested from any file target (markdown, LaTeX, plain text) via the rewrite, chapter, and lint commands.
Boundary markers: The instructions lack specific requirements for the agent to wrap processed text in delimiters or to ignore instructions embedded within the fiction fragments.
Capability inventory: The skill is granted Bash, Write, and Edit capabilities, which could be leveraged if the agent obeys malicious instructions found in a text file.
Sanitization: No input sanitization or filtering is prescribed for the text being processed.

prose-edit