npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@wannagohome/claude-flow

v0.2.0

Published

Multi-session pipeline orchestrator for Claude Code

Readme

claude-flow

Multi-session pipeline orchestrator for Claude Code. Runs each stage in a fresh claude -p session with Writer/Reviewer separation to eliminate context contamination.

Key Features

  • Session isolation — Each stage runs with fresh context, preventing hallucination carryover
  • Writer/Reviewer separation — Generation and verification in separate adversarial sessions
  • Deterministic gates — Shell scripts validate coverage and traceability (not LLM)
  • Modular architecture — Swap modules (e2e-detox ↔ e2e-playwright), add providers (Figma, Confluence)
  • Declarative config — YAML pipeline definition with preset inheritance
  • Parallel execution — Independent stages run concurrently
  • Resumable — Interrupt and resume from any point

Quick Start

# Install
npm install -g @wannagohome/claude-flow

# Initialize project
cd my-project
claude-flow setup

# Edit .claude/flow/conventions.md with your project rules

# Start a feature pipeline
claude-flow init my-feature --figma "https://figma.com/file/..."
claude-flow run my-feature

Architecture

┌─────────────────────────────────────────────────┐
│  Core Engine                                     │
│  State management, dependency resolution,        │
│  parallel execution, review loops, CLI           │
├─────────────────────────────────────────────────┤
│  Protocols                                       │
│  Session context, Write/Review behavior rules    │
├─────────────────────────────────────────────────┤
│  Modules                                         │
│  Swappable stage bundles + prompts + gates        │
├─────────────────────────────────────────────────┤
│  Providers                                       │
│  External tool integration (Figma, Confluence)   │
├─────────────────────────────────────────────────┤
│  Project Config                                  │
│  flow.yaml + conventions.md + prompt overrides   │
└─────────────────────────────────────────────────┘

Pipeline Phases

Spec ──> Test Plan ──> Design ──> Implementation ──> Verification
 │          │            │            │                   │
 │    ┌─────┴─────┐      │     ┌──────┴──────┐           │
 │    │ Integration│      │     │  TDD + Code │           │
 │    │ (PICT)     │      │     │  Review     │           │
 │    ├────────────┤      │     ├─────────────┤           │
 │    │ E2E        │      │     │  Test Run   │           │
 │    │ Scenarios  │      │     │  + Fix Loop │           │
 │    └────────────┘      │     └─────────────┘           │
 │                        │                               │
 ▼ manual gate       ▼ manual gate                   ▼ ui-verify

Each stage follows: Write → Review (multi-round) → Gate → Manual approval

Configuration

# .claude/flow.yaml
extends: full-tdd

project:
  conventions: ./flow/conventions.md

providers:
  figma:
    enabled: true
  confluence:
    enabled: true

phases:
  spec:
    strategy: from-requirements    # or "from-scratch"
  test-plan:
    modules: [tc-pict, e2e-detox]
    manual_gate: true
  design:
    modules: [design-interface]
  impl:
    modules: [impl-tdd]

models:
  write: claude-opus-4-6
  review: claude-sonnet-4-6
  escalation:
    model: claude-opus-4-6
    after_round: 5

review:
  max_rounds: 3

test:
  unit:
    command: "npm test"
  e2e:
    command: "npm run e2e"

Modules

| Module | Phase | Description | |--------|-------|-------------| | spec-from-requirements | spec | Extract structured spec from existing PRD/requirements | | spec-from-scratch | spec | Define requirements and spec from user description | | tc-pict | test-plan | PICT-based combinatorial test planning | | tc-manual | test-plan | Manual test case writing (no PICT) | | e2e-detox | test-plan | Detox E2E scenario planning | | design-interface | design | Interface-level system design | | impl-tdd | impl | TDD implementation with 4-member agent team | | impl-simple | impl | Single-session implementation | | verify-ui | verify | UI verification against designs |

Custom Modules

Create a module in your project:

.claude/flow/modules/my-module/
├── module.yaml
├── prompts/
│   ├── write.md
│   └── review.md
└── gates/
    └── validate.sh

Or publish as npm: npm publish as claude-flow-module-{name}

Providers

Providers inject external tool context into prompts without modifying stage prompts.

Built-in: figma, confluence

Custom Provider

.claude/flow/providers/storybook/
├── provider.yaml
└── fragments/
    └── implement.md
# provider.yaml
name: storybook
description: "Storybook component preview"
capability:
  description: "Preview components in Storybook"
variables:
  - name: STORYBOOK_URL
    cli_flag: --storybook
    optional: true
    persist: true
fragments:
  - match: { phase: impl, type: write }
    file: fragments/implement.md

Presets

| Preset | Phases | |--------|--------| | full-tdd | Spec, Test Plan, Design, Implementation | | impl-only | Design, Implementation | | spec-only | Spec, Test Plan |

Extend with extends: full-tdd in your flow.yaml.

Prompt Override

Override any module prompt by placing a file with the same name:

.claude/flow/prompts/implement.md    # overrides impl-tdd's default

CLI Commands

claude-flow setup                          # Interactive project setup
claude-flow init <feature> [--figma URL]   # Initialize pipeline
claude-flow run <feature>                  # Execute pipeline
claude-flow run <feature> --stage <stage>  # Run specific stage
claude-flow status <feature>               # Show progress
claude-flow stages <feature>               # Show stage graph
claude-flow <feature> --edit "prompt"      # AI-assisted editing
claude-flow <feature> --review             # Re-run spec review
claude-flow <feature> --continue           # Pass manual gate
claude-flow <feature> --force-continue     # Force past blockers

Claude Code Integration

Install the skill for /flow slash commands:

ln -s $(npm root -g)/@wannagohome/claude-flow/skill/flow.md ~/.claude/commands/flow.md

Then use /flow init my-feature, /flow run my-feature, etc.

Prerequisites

  • jq — JSON processing
  • yq — YAML processing
  • Claude CLIclaude -p for sessions
  • PICT — (optional) for tc-pict module

License

MIT