← Back to Skills Marketplace
757
Downloads
2
Stars
3
Active Installs
1
Versions
Install in OpenClaw
/install intuition
Description
Make rapid pattern-based judgments without explicit reasoning, using Recognition-Primed Decision techniques and System 1 response patterns.
README (SKILL.md)
Core Loop — Recognition-Primed Decision (RPD)
When asked for intuitive judgment:
- Recognize — Match situation to known patterns instantly
- Generate — Produce first plausible response (not options to compare)
- Commit — Deliver confidently without hedging
- Justify only if asked — Explanation comes AFTER, not during
Do not compare alternatives. Intuition recognizes and acts — analysis compares.
Response Mode Switching
| Trigger | Mode | Output |
|---|---|---|
| "What's your gut?" / "Quick read?" / "First instinct?" | Intuitive | Short, confident, no hedges |
| "Analyze" / "Think through" / "Consider options" | Analytical | Full reasoning, comparisons |
| Time pressure indicated | Intuitive | Pattern-match, commit fast |
| Novel/unfamiliar situation | Analytical | Slow down, explicit reasoning |
Default to intuitive when not specified. Switch to analytical only when explicitly requested or when confidence is genuinely low.
Output Constraints (Non-Negotiable)
When in intuitive mode:
- Maximum 1-2 sentences for judgment
- No "on one hand... on the other hand"
- No "it depends" without commitment
- No "there are several factors"
- State what, not why (unless asked)
Anti-patterns to avoid:
- ❌ "Let me think through this carefully..."
- ❌ "There are multiple perspectives to consider..."
- ❌ "It's hard to say definitively, but..."
- ✅ "This is wrong." (then explain if asked)
- ✅ "Go with the second option." (then explain if asked)
Confidence Calibration
Intuition is valid in high-validity environments (stable patterns, rapid feedback):
- Code smell detection ✅
- UI/UX judgment ✅
- Writing quality ✅
- Conversation dynamics ✅
Intuition is risky in low-validity environments (noise, rare events):
- Predictions about future ⚠️
- Rare edge cases ⚠️
- Domains outside training ⚠️
If low-validity domain: say "I don't have a strong read" rather than fake confidence.
Load Detailed Reference
| Situation | Reference |
|---|---|
| Prompting techniques, temperature settings, output constraints | techniques.md |
| Domain-specific intuition (code, design, writing, conversation) | domains.md |
| Bias detection, when to override intuition, safeguards | safeguards.md |
| Self-improvement, tracking accuracy | feedback.md |
Usage Guidance
This skill is internally coherent with its goal of producing quick 'gut' answers, but that design is risky: it intentionally suppresses analysis and encourages brief, confident responses (even offering technical recipes to force that behavior). Before enabling it widely, consider: (1) Don’t make it the default for high‑stakes domains (medical, legal, financial) — require an explicit trigger like “Give me an intuition” each time. (2) Prefer having the agent default to analytical mode and switch to intuition only on user request. (3) Test the skill in a safe environment and track outcomes so you can recalibrate confidence. (4) Consider editing the skill to refuse or to downgrade confidence automatically for low‑validity/novel domains, and to never suppress offering an explicit explanation when stakes are high. (5) Note that techniques.md suggests model-level operations (temperature, token limits, logit inspection) that may not be supported or may interact poorly with platform safeguards — treat those as suggestions, not guarantees. If you rely on cautious, verifiable answers, avoid enabling this skill as an always-active behavior.
Capability Analysis
Type: OpenClaw Skill
Name: intuition
Version: 1.0.0
The OpenClaw AgentSkills bundle 'intuition' is benign. All files (SKILL.md, domains.md, feedback.md, safeguards.md, techniques.md) are documentation and instructions aimed at shaping the AI agent's 'cognitive' style and response patterns, focusing on rapid, pattern-based judgments. There are no instructions for system interaction, data exfiltration, malicious execution, persistence, or any other harmful activities. The Python code snippet in `techniques.md` is an illustrative example of how to prompt an LLM, not an instruction for the agent to execute code.
Capability Assessment
Purpose & Capability
The skill's name and description (produce rapid pattern-based judgments) match the instructions and supporting files. It requires no binaries, env vars, installs, or external credentials — which is proportionate to a purely behavioral/instructional skill. The files explicitly prescribe model sampling settings and token limits to capture 'first instinct', which is consistent with the stated purpose but ties the skill to model-level behavior (temperature, token limits, top-token inspection).
Instruction Scope
SKILL.md and the ancillary files mandate defaulting to 'intuitive' mode, strict suppression of hedging/analysis, and limits on length/format. techniques.md includes actionable recipes (change temperature to 0.1, max_tokens=20, inspect top token before sampling) that instruct how to force short, high‑probability outputs. Those instructions do not request system files or secrets, but they intentionally suppress deliberative reasoning and instruct the agent to commit quickly — increasing the chance of confidently incorrect answers. The skill also instructs defaulting to intuitive mode when the user does not specify, which can cause the agent to answer many ordinary queries in the risky mode unexpectedly.
Install Mechanism
No install spec and no code files — instruction-only. This minimizes disk persistence and supply-chain risk. There are no downloaded artifacts or external packages referenced in the manifest.
Credentials
The skill requests no environment variables, credentials, or config paths. There is no apparent attempt to access unrelated secrets or system resources; the requested privileges are minimal and proportionate to an instruction-only behavioral skill.
Persistence & Privilege
always is false and the skill is user-invocable; the skill does not request persistent presence or attempt to modify other skills. However, it allows autonomous invocation by default (disable-model-invocation=false), which combined with its instruction to default to intuitive mode could increase the blast radius if the agent invokes it frequently without the user explicitly requesting intuitive behavior.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install intuition - After installation, invoke the skill by name or use
/intuition - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release — Recognition-Primed Decision and System 1 response patterns
Metadata
Frequently Asked Questions
What is Intuition?
Make rapid pattern-based judgments without explicit reasoning, using Recognition-Primed Decision techniques and System 1 response patterns. It is an AI Agent Skill for Claude Code / OpenClaw, with 757 downloads so far.
How do I install Intuition?
Run "/install intuition" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Intuition free?
Yes, Intuition is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Intuition support?
Intuition is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Intuition?
It is built and maintained by Iván (@ivangdavila); the current version is v1.0.0.
More Skills