By waymaker-ai
Guardrails + VibeSpec/FlowSpec feature flow + 16 skills, 4 commands, 3 agents, 3 hooks, and a built-in MCP server.
Run CoFounder safety gates on the current changeset (secrets, mock data, scope, design system, missing tests). Usage — /check or /check --diff <range>
CoFounder root command — route to feature-new, feature-implement, check, or spec-review. Usage — /cofounder <subcommand> [args]
Explain the current branch's changes as a mental model — what changed, why, blast radius — not a file list. Usage — /explain-changes [--diff <range>]
Turn a feature idea into a VibeSpec-compliant spec file. Usage — /spec <one-line idea> or /spec (interactive)
Use this agent to review a diff or PR against the repo's CoFounder VibeSpecs. Checks for secrets, mock data in prod, scope violations, design-system breaches, missing tests, and deviations from referenced specs. Returns a structured review with confidence-filtered findings — reports only high-priority issues that actually matter. Use proactively after a feature implementation or before opening a PR.
Spawnable sub-agent variant of the pattern-scout skill. Delegate to this agent when you want a separate, isolated context to find existing patterns in the repo before writing new code — searches for similar functions, components, hooks, models, and routes, then returns ranked matches with extend-vs-create recommendations. Use proactively when the user asks to "add", "build", "create", or "extract" something non-trivial. Distinct from the `cofounder-pattern-scout` skill, which is auto-invoked rather than spawned.
Use this agent when the user describes a feature idea but the scope, acceptance criteria, or data sources aren't clear. Asks 3–5 clarifying questions, searches the repo for prior art, and writes a VibeSpec-compliant feature spec to `specs/<slug>.spec.yml`. Use proactively before any non-trivial feature implementation.
Use this skill when the user asks to validate their repo against CoFounder guardrails — "run cofounder check", "verify the changes are safe", "is this PR-ready?", "check for guardrail violations". Runs typecheck, lint, tests, and CoFounder-specific rules (no secrets, no mock data in prod, no scope creep, no design-system violations) and reports a pass/fail summary.
Use this skill when the user asks to commit their staged changes, or asks "what's a good commit message for this". Writes a Conventional Commits-style message from the ACTUAL staged diff. Will NEVER commit without explicit user confirmation.
Use this skill BEFORE implementing a feature that may touch regulated data (health records, payments, EU users, financial advice, minors). Reframes the feature spec against HIPAA, GDPR/CCPA, SEC, PCI DSS, FERPA, or SOX requirements and surfaces what changes in scope, data handling, audit logging, and consent flows. Wraps `@waymakerai/aicofounder-policies` and `@waymakerai/aicofounder-compliance`.
Use this skill when the agent is hitting context limits, the prompt is too long, or "this conversation is getting expensive". Distills long inputs (long files, long histories, huge diffs, large RAG returns) into the minimum context that preserves the answer. Wraps `@waymakerai/aicofounder-context-optimizer`.
Use this skill when the user asks "which model should I use?", "how much will this cost?", or "make this cheaper". Recommends the cheapest model that's still good enough for the task, based on task class and the project's budget envelope. Wraps the cost tracking and routing in `@waymakerai/aicofounder-core`. Can also analyze recent LLM spend.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Keeps AI coding agents from wrecking your codebase. Free, open-source guardrails + spec-driven feature flow for AI-assisted development, by Waymaker AI.
CoFounder plugs into your existing stack (Vercel AI SDK, Supabase, LangChain, CrewAI, OpenClaw, Claude Code) and stops agents from committing secrets, leaking PII, shipping mock data, blowing through budgets, or refactoring things no one asked them to. It's the safety harness and coach for AI-assisted work on real products — not a magic "ship-a-startup-in-30-minutes" button.
❌ Agent commits AWS keys to repo
❌ Agent leaks patient SSN in API response
❌ Agent blows through $500 in API calls overnight
❌ Agent gives investment advice (SEC violation)
❌ Agent rewrites your auth middleware "for fun"
import { createGuard } from '@waymakerai/aicofounder-guard';
const guard = createGuard({
pii: 'redact',
injection: 'block',
budget: { limit: 50, period: 'day' },
compliance: { frameworks: ['hipaa', 'gdpr', 'sec'] }
});
const result = guard.check(userInput);
// ✅ PII redacted, injections blocked, budget enforced, compliance checked
// All in < 1ms with zero dependencies
AI frameworks love to promise:
"Ship a production-ready product in 30 minutes."
CoFounder is deliberately not that.
You can't build a real, scalable, secure product in 30 minutes — not if you care about:
CoFounder refuses that hype.
Instead, CoFounder's promise is:
CoFounder helps AI-assisted development move faster without becoming sloppier — by stopping agents from doing things no responsible engineer would do.
Not "magic production in 30 minutes," but:
CoFounder focuses on:
You still need engineers, reviews, and real operations. CoFounder just makes the AI-powered parts safer, sharper, and less painful.
CoFounder is an integration-friendly guardrail & guidance layer for AI agents working in real codebases and products.
npx claudepluginhub waymaker-ai/ranavibe --plugin cofounderAccess thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development
Complete developer toolkit for Claude Code
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Orchestrate multi-agent teams for parallel code review, hypothesis-driven debugging, and coordinated feature development using Claude Code's Agent Teams