The Agent Skills Directory

[PROMPT_INJECTION]: The skill description and body use the word 'exploit' to describe a phase of the writing process. Analysis confirms this is a creative context term ('exploration vs. exploitation') rather than a prompt injection or technical exploit attempt.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external content, creating a vulnerability surface. 1. Ingestion points: Raw markdown material read from files provided by the user. 2. Boundary markers: Absent; the instructions do not define delimiters for the input material. 3. Capability inventory: The skill has read and append access to the local filesystem for files specified by the user. 4. Sanitization: No validation or sanitization of the input text is mentioned to prevent embedded instructions from influencing the agent's behavior.

writing-shape