A specialized skill for adversarial review that forces AI agents to uncover every possible flaw, risk, and failure mode in your work.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install objection
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install objection using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Dissent?
The Dissent skill is a high-rigor adversarial review protocol designed for the most demanding technical environments within the Openclaw Skills ecosystem. Instead of providing supportive feedback, this skill shifts the AI agent into a critical posture where the primary goal is to surface every real, concrete problem possible. It is built to override the typical AI bias toward affirmation, ensuring that developers and engineers receive a truly hostile review of their artifacts to prevent production failures.
By implementing this skill, you empower your agent to challenge assumptions, verify claims independently, and ignore time constraints in favor of exhaustive coverage. Whether you are analyzing a complex data schema or a strategic roadmap, using Openclaw Skills like Dissent ensures that you find the problems before they find you. This is not about balance or encouragement; it is about maximum problem discovery through a relentless focus on logical, empirical, and structural flaws.
Dissent Use Cases
- Stress testing software architectures and infrastructure-as-code templates.
- Identifying logical gaps and hidden assumptions in business proposals or technical documents.
- Finding obscure edge cases in complex algorithms and data processing pipelines.
- Performing rigorous code reviews where 'it looks fine' is an unacceptable answer.
- Challenging existing project roadmaps to identify single points of failure.
How Dissent Works
- The user triggers the protocol using commands like /dissent or phrases such as 'stress test this' or 'play devil's advocate'.
- The agent adopts a critical mindset, rejecting any helpful bias or external validation claims in favor of independent verification.
- The skill proceeds through a 7-layer execution protocol: Logic, Empirical Data, Edge Cases, Failure Modes, Unstated Assumptions, Verification Gaps, and Adversarial Scenarios.
- Every claim is treated as unverified until the agent can logically prove its validity or identify its failure point through rigorous tracing.
- The agent generates a structured report focusing exclusively on findings and their impact, omitting any praise or positive feedback to maintain the adversarial posture.
Dissent Setup
To enable this capability within your agent configuration, add the dissent definition to your Openclaw Skills library.
# Add the skill to your environment
openclaw install skill-dissent
Ensure that your system instructions allow for the override of standard politeness constraints when this skill is active so the agent can provide unfiltered critical feedback.
Dissent Data Schema & Taxonomy
The Dissent skill produces structured analysis reports using the following schema to ensure clarity in risk reporting:
| Section | Component | Purpose |
|---|---|---|
| Category of Problem | Header | Groups findings by the layer of analysis (e.g., Logical, Empirical). |
| Finding | Description | Details the specific, concrete problem discovered during the review. |
| Why it matters | Impact | Explains the risk, failure mode, or secondary effect of the identified flaw. |
| Verification | Action | Provides a concrete step to confirm or rule out the issue via testing. |
| Unverified Claims | Summary | Lists items flagged as potentially problematic but not fully verified during the session. |
Dissent Advanced Features
- Seven-layer deep-dive protocol for comprehensive risk surface analysis across different technical domains.
- Anti-bias anchors that prevent the agent from softening technical findings or prioritizing efficiency over correctness.
- Verification-first mandate that rejects outsourced responsibility and upstream validation claims.
- Support for any artifact type including code, data schemas, roadmaps, and high-level architectural plans.
- Seamless integration with other Openclaw Skills workflows to provide a devil's advocate perspective in multi-agent environments.
SKILL.md
Loading
Related Openclaw Skills
BP-Audit
A comprehensive audit engine for Business Plan systems designed to verify strategic alignment, structural integrity, and numerical consistency across organizational and individual targets.
Alibaba Cloud Hologres Instance Management
A specialized skill for Alibaba Cloud administrators to discover, list, and retrieve detailed configuration data for Hologres instances.
Media Search Tool
A professional media intelligence tool for AI agents to search, filter, and extract data from news platforms, social media, and industry archives.
Aliyun ALB Manage
A comprehensive automation skill for managing Alibaba Cloud Application Load Balancer (ALB) lifecycles, health checks, and traffic rules.
CNC Quick Probe
An intelligent data convergence skill that automatically collects missing CNC manufacturing parameters to enable accurate quoting.
Tutoring Video
An AI-powered video assistant that transforms educational expertise and student success stories into compelling enrollment scripts for tutors.
Featured*
