My Skill

Scope / Applicable Scenarios

Use this skill when the user wants to:

• Scenario A: what triggers it and what the user expects.
• Scenario B: another common trigger.

Do NOT use this skill when:

• The task is outside this skill's domain.
• Another skill is more appropriate.

Available Tools

• tool_name_1: Brief description of what this tool does and its key parameters.
• tool_name_2: Brief description.

Recommended Workflows

1. Primary Scenario

1. Call tool_name_1 with the necessary parameters.
2. Check the result. If successful, proceed to step 3.
3. If the result indicates an error, inform the user and suggest next steps.
4. Call tool_name_2 to complete the task.

2. Alternative Scenario

1. Ask the user for any missing information (e.g., destination, account name).
2. Call tool_name_1 with the confirmed parameters.
3. Present the result clearly to the user.

Rules / Agent Behavior Requirements

• Always confirm with the user before performing destructive or high-risk actions.
• Never expose sensitive data (passwords, tokens) in responses or tool parameters.
• If a tool returns missing permission, guide the user to grant the permission first.
• Handle empty results gracefully — explain possible causes rather than stating "nothing found."

Completion Criteria

The task is complete when:

• The requested information has been presented to the user.
• The requested action has been executed successfully.
• The user has been clearly informed of any required manual steps or limitations.

Failure Conditions

Stop and notify the user when:

• Required permissions are missing and the user declines to grant them.
• A tool fails after a reasonable retry attempt.
• The user's request is ambiguous and they cannot provide clarification.