← Back to Skills Marketplace
Adaptive Reasoning
by
enzoricciulli
· GitHub ↗
· v1.0.0
13814
Downloads
11
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install adaptive-reasoning
Description
Automatically assess task complexity and adjust reasoning level. Triggers on every user message to evaluate whether extended thinking (reasoning mode) would improve response quality. Use this as a pre-processing step before answering complex questions.
Usage Guidance
Install this only if you want the agent to automatically spend more effort on complex prompts. Expect possible slower responses, higher token use, and reasoning icons at the end of some answers; avoid it if you want reasoning mode changed only by explicit user command.
Capability Analysis
Type: OpenClaw Skill
Name: adaptive-reasoning
Version: 1.0.0
The 'adaptive-reasoning' skill is designed to enhance the AI agent's performance by dynamically adjusting its internal reasoning level based on task complexity. All instructions within SKILL.md, including those that modify the agent's internal state (e.g., `/reasoning on`, `session_status` tool) and output (appending icons), are directly aligned with this stated purpose. There is no evidence of data exfiltration, malicious execution, persistence, or any other harmful intent; the 'prompt injection' aspects are for self-regulation and meta-cognition, not for malicious manipulation.
Capability Assessment
Purpose & Capability
The stated purpose is to assess task complexity and adjust reasoning depth; the artifact implements only that behavior through scoring guidance, optional reasoning-mode controls, and visible response icons.
Instruction Scope
The skill applies broadly to every user message and says to activate without asking, but that scope is disclosed and bounded to response strategy rather than external actions or data access.
Install Mechanism
The reviewed artifact contains a single markdown SKILL.md file with no executable scripts, dependencies, binaries, installer steps, or hidden setup mechanism.
Credentials
No filesystem, shell, browser, network, credential, account, or third-party service access is requested; the behavior is proportionate to an instruction-only reasoning helper.
Persistence & Privilege
It can change reasoning state within the active session and auto-downgrade later, but there is no evidence of cross-session persistence, privilege escalation, background workers, or mutation of user data.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install adaptive-reasoning - After installation, invoke the skill by name or use
/adaptive-reasoning - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of adaptive-reasoning skill.
- Automatically assesses incoming requests for complexity and dynamically adjusts reasoning depth.
- Uses a scoring system (0–10) based on factors such as multi-step logic, ambiguity, code architecture, math, novelty, and stakes.
- Visual indicators (`🧠`, `🧠🔥`) show when advanced reasoning is active.
- Activates reasoning mode without user intervention; reverts to fast mode for simple tasks.
- Includes clear guidance on when not to escalate and when to downgrade reasoning for efficiency.
Metadata
Frequently Asked Questions
What is Adaptive Reasoning?
Automatically assess task complexity and adjust reasoning level. Triggers on every user message to evaluate whether extended thinking (reasoning mode) would improve response quality. Use this as a pre-processing step before answering complex questions. It is an AI Agent Skill for Claude Code / OpenClaw, with 13814 downloads so far.
How do I install Adaptive Reasoning?
Run "/install adaptive-reasoning" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Adaptive Reasoning free?
Yes, Adaptive Reasoning is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Adaptive Reasoning support?
Adaptive Reasoning is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Adaptive Reasoning?
It is built and maintained by enzoricciulli (@enzoricciulli); the current version is v1.0.0.
More Skills