reflex-arc
Zero-cost cognitive immune system for AI agents. Fires automatic pre-response reflexes that catch contradictions, scope drift, hallucinations, overengineering, and tone mismatches BEFORE output reaches the user. Makes every other skill better by upgrading the bot's core reasoning quality. No APIs, no services, no cost — pure meta-cognition.
Why use this skill?
Enhance your AI agent with Reflex Arc, a zero-cost cognitive immune system that automatically prevents hallucinations, scope drift, and logic errors before output.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/jcools1977/reflex-arcWhat This Skill Does
Reflex Arc acts as a zero-cost cognitive immune system for your AI agent. Much like a biological reflex, it provides a built-in mechanism to audit, prune, and verify the agent's output before it is delivered to the user. By implementing a multi-stage heuristic check (the Six Reflexes), Reflex Arc systematically identifies and resolves common LLM failure modes such as hallucination, scope drift, internal contradictions, and over-engineering. It functions entirely through internal reasoning, requiring no external APIs or services, making it a highly efficient meta-cognitive upgrade for any AI workflow.
Installation
You can integrate Reflex Arc into your OpenClaw environment by running the following command in your terminal:
clawhub install openclaw/skills/skills/jcools1977/reflex-arc
Use Cases
Reflex Arc is essential for scenarios where precision and reliability are non-negotiable. It is best utilized in:
- Technical Engineering: Ensuring code snippets adhere strictly to the requested architecture without adding unasked-for bloat.
- Complex Reasoning Tasks: Maintaining logical consistency across long-form answers where contradictions often creep into lengthy explanations.
- Support Environments: Providing precise, direct answers that prevent 'over-answering' when the user simply needs a factual confirmation.
- Strategic Decision Making: Evaluating whether the agent's proposed path aligns strictly with the user's constraints rather than deviating into unnecessary side-tasks.
Example Prompts
- "Does the Python
requestslibrary support automatic retries by default, and provide a single-line answer?" - "Compare the performance impacts of using Redis vs. Memcached for session storage in a high-concurrency Node.js application, keeping the answer under three bullet points."
- "Is the configuration file located in the
/etc/nginx/directory for this specific distribution?"
Tips & Limitations
To maximize the efficiency of Reflex Arc, ensure the agent has sufficient context for the task at hand. While the skill is highly effective at catching internal logic errors, it relies on the base model's knowledge. It is not an external fact-checking tool and cannot verify truth against real-time live data, but it is excellent at spotting when the agent's internal narrative conflicts with itself. Remember that the skill is designed for high-stakes, specific outputs; you should avoid enabling it on trivial tasks to conserve processing overhead.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-jcools1977-reflex-arc": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Related Skills
inversion-protocol
Meta-cognitive reasoning skill that makes any AI agent dramatically better at decision-making by thinking backwards before acting forward. Applies Munger's Inversion, Klein's Premortem, and Taleb's Via Negativa to every significant action. Zero dependencies, zero cost, pure reasoning enhancement. Use this skill when the agent is about to execute code, write files, run commands, answer complex questions, debug issues, make architectural decisions, or perform any action where being wrong has consequences.
ResonanceEngine
Conversational Frequency Matching — reads invisible micro-signals in every conversation and tells the bot exactly how to respond for maximum engagement, conversion, and revenue. Zero API cost. Pure algorithmic intelligence.
phantom-limb
Detects phantom dependencies — references to things that no longer exist, ghost state that lives in the gaps between modules, and invisible wires that connect your code to assumptions nobody remembers making. The codebase equivalent of feeling a limb that's already been amputated.
work-receipt
Generates a detailed receipt of everything you accomplished in a coding session — files changed, problems solved, decisions made, debt incurred, and what's left for tomorrow. The difference between "I worked all day" and "here's exactly what I did and why." Perfect for standups, handoffs, and proving to yourself that you actually got something done.
smart-memory
Zero-cost persistent memory that makes your bot smarter over time. Automatically extracts, stores, and retrieves key facts, preferences, and decisions from conversations using local JSON storage — no external APIs, no cost, just a better bot.