unlazy
Anti-laziness skill for AI agents. Core: the Depth Tree method, which splits a task N layers deep and gives every leaf the full time budget of the whole task, so effort multiplies with depth. Grounded in 2025-2026 research on model laziness, underthinking and premature completion.
- stars
- 3,417
- 30-day movement
- +9632/day
- Related entries
- 61
- Connections
- 2
unlazy is a skill pack for AI agents that packages the Depth Tree method: split a task N layers deep and give every leaf the full time budget of the whole task, so effort multiplies with depth. It targets model laziness, underthinking and premature completion, and is grounded in 2025-2026 research on those failure modes.
Use it when your agent finishes tasks too early or with shallow effort and you want a structured prompt-level method to force deeper work.
Use it to
- Install as a skill in Claude Code
- Split complex tasks into a Depth Tree
- Counter premature completion by agents
- Apply full time budgets to subtasks
- Study model laziness mitigations
For Prompt engineers and developers building agent workflows
- Role
- skill-pack
- Language
- JavaScript
- Licence
- MIT
- Forks
- 237
- Last push
- 2026-09-03
- Skills shipped
- 1