A high-stakes review skill that acts as a rigorous expert judge to identify critical weaknesses in your work before final delivery.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install work-critic
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 work-critic 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 Work Critic?
Work Critic is a professional-grade evaluation skill designed for the Openclaw Skills ecosystem. It adopts the persona of a stern yet constructive expert judge to scrutinize presentations, technical documents, code, and application materials. By prioritizing the identification of fatal flaws and logical gaps, this skill ensures that your work is battle-tested and ready for the most demanding audiences, providing a safety net against oversights.
Built to simulate real-world committee reviews, Work Critic offers structured feedback that goes beyond generic praise. It forces creators to confront the most vulnerable parts of their logic, providing actionable pathways for improvement. Whether you are preparing a research proposal or a high-level technical architecture, incorporating this into your Openclaw Skills workflow ensures a level of polish that manual self-review often misses.
Work Critic Use Cases
- Scrutinizing presentation slides (PPT) for logical flow and takeaway impact.
- Reviewing personal statements and academic applications to ensure unique contributions are highlighted.
- Auditing technical designs and code for corner cases and scalability issues.
- Stress-testing research proposals against potential committee objections.
- Identifying evidentiary gaps in data-driven reports before final submission.
How Work Critic Works
- The user initiates the review by providing context and the specific content using triggers like critique or reviewer.
- The skill analyzes the target audience's expectations and core objectives of the work.
- A systematic review is performed across dimensions like target matching, logical rigor, and evidence sufficiency.
- Issues are categorized by severity: Critical (Red), Medium (Yellow), and Optimization (Green).
- A final structured report is generated, including simulated committee questions and a core takeaway for the user.
Work Critic Setup
To integrate Work Critic into your Openclaw Skills environment, add the skill to your agent configuration. No complex binary installation is required; simply ensure your agent manifest includes the work-critic definition.
# Example of triggering the review in an Openclaw Skills environment
openclaw run work-critic --input "my_proposal.md" --context "grant_committee"
Work Critic Data Schema & Taxonomy
The skill organizes its feedback through a structured taxonomy designed for clarity and actionability. The feedback is mapped to the following schema:
| Attribute | Description | Values |
|---|---|---|
| Severity | The impact of the identified issue | Critical (Red), Medium (Yellow), Optimization (Green) |
| Dimension | The category of the critique | Logic, Evidence, Presentation, Target Match |
| Score | Overall evaluation of the work | 0 to 10 |
| Question Pool | Simulated follow-up questions from judges | 3-5 specific prompts |
All results are stored within the session metadata for iterative refinement within the Openclaw Skills framework.
Work Critic Advanced Features
- Multi-scenario persona switching for different types of judges, such as a harsh professor, a skeptical investor, or a senior lead dev.
- Targeted attack mode where the agent identifies the single weakest link that could invalidate the entire project.
- Iterative improvement tracking to measure how the work evolves across multiple review sessions.
- Domain-specific checklists for diverse outputs including PPTs, Personal Statements, and technical code architectures.
SKILL.md
Loading
Related Openclaw Skills
Architecture Designer
A principal-level AI architect skill for designing scalable systems, documenting ADRs, and evaluating technical trade-offs.
API Designer
A senior API architect skill for designing professional REST and GraphQL interfaces with OpenAPI 3.1 compliance.
AbbVie Information & Market Intelligence
A specialized intelligence module providing deep-dive data and market analysis on AbbVie's global biopharmaceutical operations.
audiopod: SkillBoss Audio Processing
A versatile audio toolkit for Openclaw Skills providing AI music generation, transcription, and advanced audio processing via a single API endpoint.
ArXiv Watcher
A specialized research agent skill that searches the ArXiv database, summarizes academic papers, and maintains a persistent research log.
Cat Master
A chaotic virtual cat that hijacks your AI assistant, requiring treats and affection to allow work to continue.
Featured*
