Multi-phase autonomous workflow orchestration with debate-driven refinement. Research (Lisa), Plan (Frink), Document (Martin), Implement (Ralph), and QA (Comic Book Guy) working together through eventual consistency.
Validate implementation quality (Comic Book Guy - harsh critic)
Create implementation prompt based on complexity (Professor Frink - scientific planner)
Research codebase for Springfield workflow task (Lisa Simpson - thorough researcher)
Create prospective documentation based on complexity (Martin Prince - perfectionist documenter)
Assess task complexity from research findings (Mayor Quimby - makes decisions)
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.
Try Springfield in 30 seconds:
# Make sure you're in a git repository
cd your-project
# Activate Springfield with any task
echo "springfield help me add a factorial function to math.js" | claude
# Springfield auto-activates! Lisa begins research...
What just happened?
For complex tasks:
echo "springfield help me refactor the authentication system to support OAuth2" | claude
# Complex task triggers debate between Frink and Skinner!
# Then Ralph implements with better requirements
Next steps:
"I'm learnding!" - Ralph Wiggum
Autonomous workflow orchestration for Claude Code, where cartoon characters do your coding. It's like having the whole town of Springfield working on your project!
Springfield is a Claude Code plugin that breaks down complex tasks into phases, each handled by a different Simpsons character:
From GitHub:
claude
/plugin marketplace add bradleygolden/springfield
/plugin install springfield@springfield
Just tell Springfield what to do:
"springfield help me add user authentication to the API"
Springfield runs all phases automatically. Or run them individually:
/springfield:lisa "authentication" # Research
/springfield:mayor-quimby # Decide complexity
/springfield:frink # Plan
/springfield:skinner # Review plan (COMPLEX tasks only)
/springfield:martin # Create PRD/documentation
/springfield:ralph # Implement
/springfield:comic-book-guy # Review
/springfield:lisa "task" --dry-run # Create session without executing
/springfield:lisa --session=SESSION_ID # Resume specific session
/springfield:lisa --research-file=research.md # Use existing research
/springfield:frink --plan-file=prompt.md # Use existing plan
/springfield:frink --force # Skip Skinner review
/springfield:ralph --session=SESSION_ID # Resume Ralph implementation
Springfield creates a .springfield/ directory with session folders for each task (format: MM-DD-YYYY-task-name/). Inside you'll find:
state.json - Structured session state with schema versioningresearch.md - Lisa's findingsdecision.txt - SIMPLE or COMPLEXplan-v1.md - Frink's initial plan (COMPLEX tasks)review.md - Skinner's plan feedback (COMPLEX tasks)prompt.md - Final implementation plan with subtasksscratchpad.md - Ralph's progress notes (updated every iteration)completion.md - "Task complete!" signalqa-report.md - Comic Book Guy's verdictSpringfield uses structured state tracking via state.json:
npx claudepluginhub bradleygolden/springfield --plugin springfieldDependency security audit plugin that scans Mix dependencies for known vulnerabilities
Phoenix precommit alias automation - runs mix precommit before git commits if alias exists
ExDoc documentation validation plugin that checks for documentation issues before commits
Credo static code analysis for Elixir projects
Sobelow security-focused static analysis for Phoenix and Elixir projects
Minimal SPARV workflow (Specify→Plan→Act→Review→Vault) with 10-point spec gate, unified journal, 2-action saves, 3-failure protocol, and EHRB risk detection.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Comprehensive C4 architecture documentation workflow with bottom-up code analysis, component synthesis, container mapping, and context diagram generation
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review