The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external content provided by users for editorial review and improvement.
Ingestion points: The skill instructions in SKILL.md identify 'The copy to be reviewed (pasted in full or linked)' and performance data as primary inputs.
Boundary markers: There are no explicit delimiters or specific instructions for the agent to treat ingested text strictly as data, increasing the risk of the agent following instructions embedded within the user-provided content.
Capability inventory: The skill body is composed of text processing instructions; however, the agent's execution environment may provide tools that could be targeted if the agent is coerced by malicious input.
Sanitization: No instructions or mechanisms are provided to sanitize or validate the content ingested from external sources.
[NO_CODE]: The skill consists entirely of instructional markdown and does not include any executable scripts, binaries, or configuration files that could carry code-based payloads.

copy-editing