The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes output/DRAFT.md as its primary input. This text is untrusted and could contain malicious instructions designed to influence the agent's behavior during the polishing process.\n
Ingestion points: The output/DRAFT.md file is read and processed in scripts/run.py and used as context by the agent.\n
Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands within the draft text.\n
Capability inventory: The associated script scripts/run.py has the capability to read and write files within the workspace using atomic_write_text. It does not demonstrate network or shell execution capabilities.\n
Sanitization: The skill does not perform sanitization or filtering of the input text before processing or rewriting it.

draft-polisher