The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts/validate.sh script executes a local bash script (heal.sh) located in a relative path within the repository structure. This is used for checking the skill's integrity and adherence to formatting standards.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, specifically proposed process work and outcome descriptions. It lacks explicit boundary markers or sanitization logic for these inputs. However, the skill's defined capabilities are limited to producing text-based judgments and decisions, and it does not have access to high-risk tools like network operations or persistent file modifications that would make this surface highly exploitable.

anti-ceremony