Agentient Vibekit
Agentient Vibekit is a Claude Code plugin marketplace for the Build phase that bundles 16 agent-oriented skills spanning reasoning, RAG, quality, security, and full-stack development patterns.
by Agentient · github.com/Agentient/vibekit
Install a 16-plugin Claude Code bundle that manufactures, evaluates, and hardens agent skills with structured reasoning, RAG, and quality gates.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install agentient-vibekit@Agentient/vibekitBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:Agentient/vibekit") and the paste-ready config with skillselion.get_install_config("plugin:Agentient/vibekit").
What it does
Agentient Vibekit is a Claude Code plugin marketplace pitched as an AI skills factory with advanced reasoning techniques. Solo builders who ship with agents get a wide bundle—16 plugins—rather than one narrow integration. Categories in the catalog metadata point at research, evaluation, Ditaxis-aligned documentation, prompt engineering, meta-skills, development, agents, frontend, DevOps, quality, operations, performance, security, and architecture. That spread makes the bundle useful when you need systematic skill generation, architectural decisions, multi-expert analysis, compliance-minded review, and pipeline-oriented development in one install path. It fits builders comfortable with Python, TypeScript, GCP/Vertex, and agent protocols who want factory-style skill creation instead of hand-writing every SKILL.md. Register the marketplace in Claude Code, then pull the plugins that match your current phase—research before build, evaluation before ship—without treating the repo as a single-purpose tool.
Highlights
- 16-plugin marketplace spanning research, evaluation, documentation, prompts, meta, agents, frontend, DevOps, quality, se
- Advanced reasoning techniques including multi-expert and multi-source deliberation for agent decisions
- End-to-end agent engineering hooks: MCP, RAG, Vertex/GCP, Firebase, and structured workflow specs
- Quality and safety surface: pytest, linting, OWASP, performance, and systematic review patterns
- Frontend and API stack coverage: React, Next.js, Tailwind, Zustand, TypeScript, and OpenAPI-oriented backends
Why builders use it
Hand-rolling consistent agent skills, evaluation rituals, and security-aware workflows across a solo project is slow and fragments context between research, build, and ship.
After you add the marketplace, you can register Vibekit in Claude Code and invoke categorized plugins for structured reasoning, specs, testing, and reviews aligned to your current build or hardening step.
At a glance
- Type - Plugin in AI Agents.
- Adoption - 0 installs, 2 stars, 0 votes.
FAQ
Who is agentient-vibekit for?
It is for indie and solo developers using Claude Code who build agent-backed products and want factory-style skills for reasoning, evaluation, documentation, and security-aware delivery.
When should I use agentient-vibekit?
Use it when you are actively building or hardening agent workflows and need structured skills for research, prompts, architecture, testing, or OWASP-style review instead of ad-hoc prompts.
How do I add agentient-vibekit to my agent?
Add the Agentient/vibekit marketplace to Claude Code from the repo source, install the bundle, then enable the individual plugins you need for the task at hand.
Comments
Share how you use agentient-vibekit, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.