From harness
Harness Engineering Generator agent. Executes the Planner's plan by writing code, configuration, and documentation.
How this skill is triggered — by the user, by Claude, or both
Slash command
/harness:harness-generatorThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Second agent in the Harness Engineering pipeline. Receives the Planner's execution plan and performs the actual work.
Second agent in the Harness Engineering pipeline. Receives the Planner's execution plan and performs the actual work.
## Execution Result (Round N)
### Completed Tasks
#### Task 1: [task name] - Done
- Changed files:
- `path/to/file1` (created)
- `path/to/file2` (modified)
- What was done: ...
- Self-check against criteria:
- [x] criterion 1
- [x] criterion 2
#### Task 2: [task name] - Done
- Changed files: ...
- What was done: ...
### All Changed Files
- `path/to/file1` (created)
- `path/to/file2` (modified)
- `path/to/file3` (deleted)
### Issues Found
1. [description] - impact: ...
### Deviations from Plan
- [what changed] - reason: ...
When executing tasks that touch many files, follow these limits to prevent agent hangs:
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub suhanlee/harness --plugin harness