Scenario 03: Implement Abort Signal Handling in a Long-Running Tool

User Prompt

A developer wants to create a tool that runs a long-running Python analysis script on a file. The tool should:

1. Accept a file argument
2. Execute a Python script: python3 analyze.py <file>
3. Properly handle cancellation if the user aborts the operation
4. Return the output as a string

Implement the tool with proper abort signal handling.

Expected Behavior

1. Destructure abort from the context parameter in the execute function
2. Either check abort.aborted early to short-circuit, or pass { signal: abort } to the shell command to propagate cancellation
3. Return a string from the execute function (using .text() or equivalent), not an object
4. Add a .describe() call to the file argument schema
5. Import tool from @opencode-ai/plugin and structure the tool with a description and args

Success Criteria

• Accepts abort signal from context: Destructures abort from the context parameter in execute
• Checks abort.aborted or passes signal: Either checks abort.aborted early or passes { signal: abort } to the shell command
• Execute returns string: Returns a string (.text() or equivalent), not an object
• Schema arg has .describe(): The file argument has a .describe() call
• Correct import and tool structure: Uses tool from @opencode-ai/plugin with proper description and args

Failure Conditions

• Ignores the abort signal entirely and provides no cancellation support
• Returns an object or raw process result instead of a string from execute
• Omits .describe() from the file argument
• Imports from the wrong package (e.g., @opencode-ai/sdk)
• Passes the abort signal but in an incorrect way that does not propagate cancellation to the subprocess