Past battle · 2025-02-06 UTC
Skills Showdown — February 6, 2025
From the Skills category. 31 marks placed across 9 fighters. Brand Kit (Grade A) took the crown.
Final standings
The line-up
The fighters
Profiles of every tool that competed in this battle, ranked by their final score.
Brand Kit (Grade A)
Security-tested data-ai skill for Claude AI. Grade A. Build a portable brand-kit.json for the user's product — colors, typography, voice — that other skills (video, ad creative, landing page) can cons
The Brand Kit skill for Claude AI generates a portable brand-kit.json file that describes a user's product, including colors, typography, voice, and optional logo. This file can be consumed by other skills, such as video, ad creative, and landing page. The skill accommodates different scenarios, allowing users to extract branding information from their website's source code, a public URL, or a logo image. It provides multiple discovery paths to create the brand kit, prioritizing methods that yield sufficient information. The output is a JSON file containing essential branding details, such as colors, typography, and voice guidelines. The skill ensures consistency across various marketing materials by providing a standardized brand kit.
Criteria breakdown
- Extract branding information from various sources (site URL, repo path, logo path)
- Generate a portable brand-kit.json file
- Include essential branding details (colors, typography, voice)
- Optional logo inclusion
Kyosei (Grade A)
Security-tested data-ai skill for Claude AI. Grade A. Code review for PRs or local changes. Covers code quality, dependency updates, performance, test coverage, documentation accuracy, and security. U
Kyosei (Grade A) is a security-tested data-ai skill for Claude AI, designed for comprehensive code reviews. It assesses code quality, dependency updates, performance, test coverage, documentation accuracy, and security. The tool is suitable for reviewing Pull Requests (PRs), checking code quality, and running thorough code evaluations. Kyosei operates by utilizing multiple specialized sub-agents in parallel to provide an all-encompassing review. These sub-agents focus on specific areas such as code quality, dependencies, documentation, performance, security, and tests. The tool can be used in GitHub output mode, where results are posted as inline comments on GitHub PRs, or in local output mode, where results are directly printed to the terminal. It also supports simplified reviews for cases where the changeset between the current and previous review is minimal or empty.
Criteria breakdown
- Code quality review
- Dependency updates assessment
- Performance evaluation
- Test coverage analysis
- Documentation accuracy check
- Security review
Persona Dev (Grade D)
Security-tested data-ai skill for Claude AI. Grade D. This skill should be used when the user asks to "create a new persona", "build me a persona", "add a skill to an existing persona", "update a pers
Persona Dev is a security-tested data-ai skill for Claude AI, used for creating, updating, and improving personas. It handles tasks such as building new personas, adding skills to existing ones, and setting up hooks or self-improvement. Personas are stored in git-tracked directories with sandbox isolation, containing settings, skills, and personal data. The skill guides users through a structured discovery process to understand the persona's domain, personality, and workflows.
Criteria breakdown
- Persona creation
- Skill addition
- Workflow setup
- Hooks and self-improvement
- Git integration
Api Docs (Grade A)
Security-tested development skill for Claude AI. Grade A. Workflow for creating comprehensive API documentation. Use when the user needs to document API endpoints.
Api Docs (Grade A) is a workflow for creating comprehensive API documentation. It uses when the user needs to document API endpoints. The process involves inventorying endpoints, documenting each endpoint, creating an OpenAPI spec, and generating documentation using Swagger UI or Redoc. The user must have API endpoints available and access to OpenAPI/Swagger tooling. Steps to document API endpoints include listing all endpoints, grouping by resource/entity, and noting HTTP method and path. This skill also covers validation and auto-updating of documentation in CI/CD, ensuring that all endpoints are accurately documented with correct examples and error responses.
Criteria breakdown
- Endpoint inventory and grouping
- Detailed endpoint documentation
- OpenAPI specification creation
- Interactive documentation generation
- Validation and review checklist
Repo Analyze (Grade A)
Security-tested data-ai skill for Claude AI. Grade A. One-call cold-start orientation for an unfamiliar repo. Returns a structured map — file tree (compact per-folder summary by default, full per-file
Repo Analyze is a security-tested data-ai skill for Claude AI that provides a one-call cold-start orientation for an unfamiliar repository. It returns a structured map of the file tree, language mix, framework and build-tool detection, top-level dependencies, entrypoints, and README highlights. This tool is designed to save time by reducing the need for multiple Read, Glob, and Grep calls when exploring a new repository. It's stateless, doesn't require network access, and is implemented in standard library Python.
Criteria breakdown
- File tree summarization (compact or full)
- Language mix analysis
- Framework and build-tool detection
- Top-level dependency listing
- Entrypoint identification
- README highlights extraction
Shogun Bloom Config (Grade A)
Security-tested data-ai skill for Claude AI. Grade A. Interactive wizard: guided questions with multiple-choice options about subscriptions, then outputs a ready-to-paste capability_tiers YAML + fixed
Shogun Bloom Config (Grade A) is a security-tested data-ai skill for Claude AI. It's an interactive wizard that guides users through a series of multi-choice questions about subscriptions, and outputs a ready-to-paste capability_tiers YAML + fixed agent model assignments. The questions cover topics such as Claude plans, ChatGPT plans, and rate limit preferences. The output is a configuration that can be pasted into a settings file. The tool is designed to help users set up their Claude and ChatGPT subscriptions for optimal performance. It's available at https://github.com/yohey-w/multi-agent-shogun/blob/main/skills/shogun-bloom-config/SKILL.md.
Criteria breakdown
- Interactive guided questions
- Capability_tiers YAML output
- Fixed agent model assignments
- Cost group declarations
- Gap warnings for unsupported Bloom levels
Frontend Design (Grade A)
Security-tested development skill for Claude AI. Grade A. Build web interfaces with genuine design quality, not AI slop. Use for any frontend work - landing pages, web apps, dashboards, admin panels,
Frontend Design is a security-tested development skill for Claude AI that helps build web interfaces with genuine design quality. It's used for frontend work such as landing pages, web apps, dashboards, admin panels, and components. The skill detects existing design systems and respects them, covering composition, typography, color, motion, and copy. It verifies results via screenshots before declaring done, ensuring production-grade frontend interfaces that avoid generic AI aesthetics. The skill follows a hierarchy of authority, prioritizing existing design systems and user instructions. It works by detecting context, planning the design, building with intention, and verifying visually.
Criteria breakdown
- Context detection
- Design planning
- Visual verification
- Respect for existing design systems
- Opinionated guidance for typography, color, motion
Visual Explainer (Grade A)
Security-tested data-ai skill for Claude AI. Grade A. Generate self-contained HTML pages that visually explain systems, data stories, investigations, editorial workflows, and code changes. Use when th
Visual Explainer is a security-tested data-ai skill for Claude AI that generates self-contained HTML pages to visually explain systems, data stories, investigations, editorial workflows, and code changes. It is designed to be used when a user requests a diagram, architecture overview, diff review, plan review, project recap, source map, comparison table, timeline, or any visual explanation of technical or editorial concepts. The skill can also proactively render complex ASCII tables as styled HTML pages. It was adapted from nicobailon/visual-explainer with journalism, newsroom, and academic design sensibilities.
Criteria breakdown
- Generate self-contained HTML pages for technical diagrams and editorial visualizations
- Proactive table rendering for complex data
- Detailed prompt templates for various use cases
- Available commands for diff review, plan review, project recap, and more
- Adapted with specific design sensibilities for journalism, newsroom, and academia
Public (Grade A)
Security-tested data-ai skill for Claude AI. Grade A. Persistent cloud memory for OpenClaw agents. Use when users say: - "install mem9" - "setup memory" - "add memory plugin" - "openclaw memory" - "m
mem9 is a security-tested data-ai skill for Claude AI, providing persistent cloud memory for OpenClaw agents. It allows users to store and recall long-term memory across sessions, devices, or restarts. The plugin connects to a cloud memory space identified by an API key, which is used to reconnect the user to the same memory space. mem9 is suitable for users who want to persist or manage long-term memory, such as remembering user preferences, project context, or important decisions. The skill should not be used for temporary conversation context or one-off tasks that do not require long-term memory.
Criteria breakdown
- Persistent cloud memory
- API key authentication
- Import existing memory files
- Multi-agent workflow support