Tree Of Thoughts

Multi-path reasoning for complex problems. Explore multiple solution branches → Evaluate each → Select optimal path. Use for: difficult decisions, creative p...

This package appears to be mostly documentation: the README (SKILL.md) describes a working Tree-of-Thoughts tool, but the runtime code is just a wrapper that tries to load a missing implementation and PLANNED.md says the feature is not implemented. It is unlikely to perform the advertised functions as shipped. If you need this capability, ask the publisher for the missing implementation (tree_of_thoughts.py) or a working release; avoid relying on this skill in production until the implementation is present and you can review it. If you still want to try it, run it in a sandboxed environment and watch logs for the wrapper's import error. Additional information that would change this assessment: inclusion of the full implementation file, or a clear reason why the wrapper should load an external module (and verification of that module's contents).

Type: OpenClaw Skill Name: tree-of-thoughts Version: 2.0.0 The skill is explicitly marked as 'PLANNED - Not yet implemented' in PLANNED.md. However, unified_wrapper.py attempts to dynamically load a non-existent 'tree_of_thoughts.py' and passes unsanitized user input ('goal') directly to its methods (do, execute, run). This design pattern creates a significant vulnerability for command injection or remote code execution if the 'tree_of_thoughts.py' file were present and processed the 'goal' argument in an unsafe manner (e.g., via os.system or subprocess.run with shell=True). While currently inert due to the missing implementation, the wrapper's design exhibits a high-risk vulnerability pattern.