The Agent Skills Directory

[SAFE]: No security issues detected. The skill's instructions align with its stated purpose of infrastructure management and health monitoring.
[COMMAND_EXECUTION]: The skill facilitates sensitive administrative actions, including the creation, revocation, and deletion of registration tokens through MCP tool calls (harness_create, harness_execute, harness_delete). This is expected behavior for the skill's purpose.
[PROMPT_INJECTION]: Identifies a surface for indirect prompt injection as the skill processes data from the Harness environment that could be influenced by external actors.
Ingestion points: Output from harness_list and harness_get tools in SKILL.md (e.g., delegate names, token descriptions).
Boundary markers: None defined in the instructions.
Capability inventory: Creation, revocation, and deletion operations via MCP tools as specified in SKILL.md.
Sanitization: No explicit validation or escaping of retrieved resource names or data content is performed before processing.

manage-delegates