The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses specific Harness MCP tools (harness_list, harness_get, harness_create) to interact with the Harness API. These are platform-integrated tools used for their intended purpose of managing security exemptions.
[SAFE]: The skill implements strong validation logic, such as enforcing mutual exclusivity between pipeline_id and target_id, and adhering to a 100-item bulk limit to prevent API errors.
[SAFE]: Instructions explicitly warn against inventing reasons or defaulting types for security exemptions, requiring user input for critical fields.
[SAFE]: Data handling is restricted to the Harness platform (Project/Org scope) and does not involve external network calls to non-vendor domains.

exempt-vuln