The Agent Skills Directory

[SAFE]: The skill is a documentation and instruction set for an AI agent to help it generate Rego (Open Policy Agent) code for Harness governance. It uses a specific MCP server, harness-mcp-v2, to execute creation and listing operations.
[COMMAND_EXECUTION]: The skill instructions involve calling MCP tools such as harness_create and harness_list to manage OPA policies. These are standard platform operations for managing Harness resources and do not involve arbitrary shell execution on the host system.
[PROMPT_INJECTION]: No attempts to override safety guidelines or extract system prompts were detected. The instructions are focused on guiding the agent to produce correct Rego syntax and follow Harness-specific schemas.
[DATA_EXFILTRATION]: No evidence of unauthorized data collection or exfiltration was found. The skill interacts solely with the specified Harness MCP server for its intended purpose of policy management.
[EXTERNAL_DOWNLOADS]: The skill does not perform any external downloads or script executions from remote sources. All reference material is contained within the skill's file structure.
[OBFUSCATION]: No obfuscated code, Base64 strings, or hidden content was found in the analyzed files.

create-policy