Wondel.ai Agent Skills
62 agent skills from bestselling books: 50 expert frameworks + 12 guided journeys for Claude Code, Codex, and more.
Evidence shows installation instructions and plugin marketplace, but no explicit least privilege or user confirmation mechanisms. Data flow transparency and sensitive data handling not mentioned. Dependency security: uses GitHub Actions and standard scripts, but no dependency audit. External effects: skills may modify project files, but not explicitly stated. Rollback not mentioned. Source attribution: README lists books and authors for each skill, but not verified.
Self-consistency: README and install instructions are consistent, but no tests or verification provided. Dependency availability: relies on external services like skills.sh and GitHub, but no guarantees. Failure messages: no error handling or user guidance provided.
Audience and scenarios: README clearly lists target users and scenarios. Capability boundaries: limitations of skills not explicitly stated. Trigger precision: skill names and descriptions are clear, but no trigger conditions provided. Environment fit: supports multiple agents and installation methods, but no detailed configuration.
Information architecture: README is well-structured with TOC and tables. Install notes are detailed. Naming stability: skill names are consistent. Examples and FAQ: README provides install examples, but no FAQ. Known limitations not mentioned. License is MIT. Versioning and changelog: GitHub Actions syncs versions, but no CHANGELOG. Maintenance responsibility: GitHub Actions auto-sync, but maintainer not explicitly stated.
Output usability: skills provide structured frameworks, but no actual output examples. Marginal value: skills based on bestsellers may offer unique value, but not verified. Cost-benefit: low installation and usage cost, but no performance or effectiveness data.
Claim traceability: README lists source books and authors for each skill. Cross-source corroboration: no independent verification provided. Fact-inference separation: not explicitly distinguished.
- No least privilege or user confirmation mechanisms; skills may directly modify project files.
- No data flow transparency or sensitive data handling; use with caution.
- No rollback mechanism; changes may be irreversible.
- No known limitations or FAQ; users may encounter unexpected issues.
- No tests or verification; skill effectiveness unverified.
What does this agent do, and when should you use it?
Wondel.ai Agent Skills delivers 62 agent skills covering business, marketing, UX, and coding, each encapsulating expert frameworks from bestselling books. It includes 50 individual skills (e.g., JTBD, CRO methodology, Refactoring UI) and 12 metaskills—guided journeys that orchestrate multiple skills step-by-step to create, improve, or grow a business, website, or app, and to improve code quality or design architecture. Skills follow the open SKILL.md standard, compatible with agentskills.io agents like Claude Code, Codex, Cursor, and others. Installation is via Claude Code plugin marketplace or skills.sh CLI.
The repository provides installable skill packages for agents like Claude Code and Codex, containing skill definitions, prompts, and guidelines. Users install via plugin marketplace commands or npx skills add. Each skill gives agents domain-specific expertise in areas like marketing, UX, coding, etc. Metaskills guide agents through phased questions, maintaining state in the project's docs/ folder. The repo also includes a Codex plugin marketplace configuration.
- A founder uses the JTBD skill to discover customers' real needs when designing a new product.
- A marketer rewrites website copy with the StoryBrand skill to clarify messaging.
- A developer refactors code using the Clean Code skill for maintainability.
- A PM validates an idea in 5 days with the Design Sprint skill.
- A UX designer improves visual hierarchy using Refactoring UI.
- A team uses the improve-business metaskill to diagnose and fix a plateaued business.
What are this agent's strengths and limitations?
- Curated frameworks from 50 bestselling books provide deep domain knowledge.
- Metaskills maintain state across sessions for extended projects.
- Multiple installation paths (plugin marketplace, skills.sh) support major agent platforms.
- Skills are organized into collections, allowing selective installation to avoid bloat.
- Dependent on specific agent ecosystems (e.g., Claude Code, Codex), limiting portability.
- Skill quality relies on book frameworks, which may require validation for your context.
- Installation requires CLI commands, potentially challenging for non-technical users.
- Metaskill state stored in project docs/ may not suit multi-environment collaboration.
How do you install or deploy this agent?
Via Claude Code: run /plugin marketplace add wondelai/skills then install collections like /plugin install product-strategy@wondelai-skills. Via skills.sh: run npx skills add wondelai/skills --all --global or individual skills like npx skills add wondelai/skills/jobs-to-be-done --global.
How do you use this agent?
After installation, invoke skills in Claude Code or Codex by referencing them in prompts, e.g., "Use jobs-to-be-done skill to write customer interview questions." Metaskills ask phased questions and save progress to docs/.