TÂCHES Windsurf Cascade Resources

A Windsurf-compatible port of the TÂCHES Claude Code Resources — the same workflows and skills, adapted for Windsurf Cascade.

Philosophy

When you use a tool like Windsurf Cascade, it's your responsibility to assume everything is possible.

I built these tools using that mindset.

Dream big. Happy building.

— TÂCHES

What's Inside

Workflows (39 total) - Slash workflows that expand into structured Cascade sessions

• Meta-Prompting: Separate planning from execution with staged prompts
• Todo Management: Capture context mid-work, resume later with full state
• Thinking Models: Mental frameworks (first principles, inversion, 80/20, etc.)
• Research: Deep dives, competitive analysis, feasibility, landscape mapping
• Deep Analysis: Systematic debugging methodology with evidence and hypothesis testing

Skills (10 total) - Autonomous workflows that research, generate, and self-heal

• Create Plans: Hierarchical project planning for solo developer + Cascade workflows
• Create MCP Servers: Build MCP servers for Windsurf integrations (Python/TypeScript)
• Create Agent Skills: Build new skills by describing what you want
• Create Meta-Prompts: Generate staged workflow prompts with dependency detection
• Create Slash Commands: Build custom commands with proper structure
• Create Subagents: Build specialized agent instances for isolated contexts
• Create Hooks: Build event-driven automation
• Debug Like Expert: Systematic debugging with evidence gathering and hypothesis testing
• Setup Ralph: Set up Geoffrey Huntley's Ralph Wiggum autonomous coding loop

Installation

Installs to ~/.codeium/windsurf/ — globally available in every project. Restart Windsurf after install/update/uninstall.

Windows (PowerShell) — one-liner

irm https://raw.githubusercontent.com/slicingmelon/taches-windsurf-cascade-resources/main/install.ps1 | iex

Update:

irm https://raw.githubusercontent.com/slicingmelon/taches-windsurf-cascade-resources/main/install.ps1 -OutFile install.ps1; .\install.ps1 -Action update

Uninstall:

irm https://raw.githubusercontent.com/slicingmelon/taches-windsurf-cascade-resources/main/install.ps1 -OutFile install.ps1; .\install.ps1 -Action uninstall

---

macOS / Linux (Bash) — one-liner

curl -fsSL https://raw.githubusercontent.com/slicingmelon/taches-windsurf-cascade-resources/main/install.sh | bash

Update:

curl -fsSL https://raw.githubusercontent.com/slicingmelon/taches-windsurf-cascade-resources/main/install.sh | bash -s -- update

Uninstall:

curl -fsSL https://raw.githubusercontent.com/slicingmelon/taches-windsurf-cascade-resources/main/install.sh | bash -s -- uninstall

---

What the installer does

• Downloads all windsurf/workflows/, windsurf/skills/, and windsurf/rules/ files directly from GitHub
• Places them in ~/.codeium/windsurf/global_workflows/, ~/.codeium/windsurf/skills/, and ~/.codeium/windsurf/global_rules/
• Saves a manifest at ~/.codeium/windsurf/taches-install-manifest.json for clean uninstall/update
• update overwrites all files; uninstall removes only files tracked in the manifest

The installer scripts are install.ps1 and install.sh at the repo root.

---

Option: Clone and use directly (single project)

Clone and open in Windsurf — the installer runs locally from the cloned folder.

git clone https://github.com/slicingmelon/taches-windsurf-cascade-resources.git

Workflows install to ~/.codeium/windsurf/global_workflows/. Skills install to ~/.codeium/windsurf/skills/. Project-specific data (prompts, todos) lives in each project's working directory.

Workflows

Meta-Prompting

Separate analysis from execution. Describe what you want in natural language, Cascade generates a rigorous prompt, then runs it in a fresh context.

• /create-prompt - Generate optimized prompts with XML structure
• /run-prompt - Execute saved prompts in delegated sub-task contexts

Todo Management

Capture ideas mid-conversation without derailing current work. Resume later with full context intact.

• /add-to-todos - Capture tasks with full context
• /check-todos - Resume work on captured tasks

Context Handoff

Create structured handoff documents to continue work in a fresh context. Reference with @whats-next.md to resume seamlessly.

• /whats-next - Create handoff document for fresh context

Create Extensions

Workflows that invoke the skills below.

• /create-agent-skill - Create a new skill
• /create-meta-prompt - Create staged workflow prompts
• /create-slash-command - Create a new slash command
• /create-subagent - Create a new subagent
• /create-hook - Create a new hook
• /create-plan - Create a hierarchical project plan

Audit Extensions

Invoke auditor skill methodology.

• /audit-skill - Audit skill for best practices
• /audit-slash-command - Audit command for best practices
• /audit-subagent - Audit subagent for best practices

Self-Improvement

• /heal-skill - Fix skills based on execution issues

Thinking Models

Apply mental frameworks to decisions and problems.

• /pareto - Apply 80/20 rule to focus on what matters
• /first-principles - Break down to fundamentals and rebuild
• /inversion - Solve backwards (what guarantees failure?)
• /second-order - Think through consequences of consequences
• /5-whys - Drill to root cause
• /occams-razor - Find simplest explanation
• /one-thing - Identify highest-leverage action
• /swot - Map strengths, weaknesses, opportunities, threats
• /eisenhower-matrix - Prioritize by urgent/important
• /10-10-10 - Evaluate across time horizons
• /opportunity-cost - Analyze what you give up
• /via-negativa - Improve by removing

Research

Systematic research with structured output saved to artifacts/research/.

• /competitive - Research competitors: who else does this, how, strengths/weaknesses
• /deep-dive - Comprehensive investigation of a topic with sources
• /feasibility - Reality check: can we actually do this with our constraints?
• /history - Research what's been tried before, lessons learned
• /landscape - Map the space: tools, players, trends, gaps
• /open-source - Find open-source libraries and tools that solve this
• /options - Compare multiple options side-by-side with recommendation
• /technical - Research how to implement something: approaches, libraries, tradeoffs

Deep Analysis

Systematic debugging with methodical investigation.

• /debug - Apply expert debugging methodology to investigate issues

---

Skills

Create Plans

Hierarchical project planning optimized for solo developer + Cascade. Create executable plans that Cascade runs, not enterprise documentation that sits unused.

PLAN.md IS the prompt - not documentation that gets transformed later. Brief → Roadmap → Research (if needed) → PLAN.md → Execute → SUMMARY.md.

Commands: /create-plan (invoke skill), /run-plan (execute PLAN.md with intelligent segmentation)

Create Agent Skills

Build skills by describing what you want. Asks clarifying questions, researches APIs if needed, and generates properly structured skill files.

When things don't work perfectly, /heal-skill analyzes what went wrong and updates the skill based on what actually worked.

Commands: /create-agent-skill, /heal-skill, /audit-skill

Create Meta-Prompts

Builds prompts with structured outputs (research.md, plan.md) that subsequent prompts can parse. Adds automatic dependency detection to chain research → plan → implement workflows.

Commands: /create-meta-prompt

Create Slash Commands

Build commands that expand into full prompts when invoked. Describe the command you want, get proper configuration with arguments and dynamic context loading.

Commands: /create-slash-command, /audit-slash-command

Create Subagents

Build specialized agent instances that run in isolated contexts. Describe the agent's purpose, get optimized system prompts with the right tool access and orchestration patterns.

Commands: /create-subagent, /audit-subagent

Create Hooks

Build event-driven automation that triggers on tool calls, session events, or prompt submissions. Describe what you want to automate, get working Cascade hook configurations.

Commands: /create-hook

Create MCP Servers

Build Model Context Protocol (MCP) servers that expose tools, resources, and prompts to Cascade. Supports Python and TypeScript implementations.

Debug Like Expert

Deep analysis debugging mode for complex issues. Activates methodical investigation protocol with evidence gathering, hypothesis testing, and rigorous verification.

Commands: /debug

Setup Ralph

Set up Geoffrey Huntley's Ralph Wiggum autonomous coding loop. Ralph is an autonomous AI coding methodology that uses iterative loops with task selection, execution, and validation.

Three phases: Planning (gap analysis → TODO list), Building (implement one task, validate, commit), Observation (you engineer the environment).

Commands: /setup-ralph

The Pirate Bay

Search The Pirate Bay for torrents and extract magnet links via the apibay.org JSON API. Supports searching by keyword, browsing top torrents by category, and filtering by seeders.

Commands: invoked automatically when asked to search for torrents or find magnet links

---

Recommended Workflow

For building projects: Use /create-plan to invoke the Create Plans skill. After planning, use /run-plan to execute phases with intelligent segmentation.

For research: Use the /research/* workflows. Each saves structured output to artifacts/research/ in your working directory.

For decisions: Use the thinking model workflows (/pareto, /first-principles, /inversion, etc.) to apply mental frameworks to any problem or discussion.

Other tools: /create-prompt + /run-prompt for custom Cascade-to-Cascade pipelines.

---

More resources coming soon.

---

Changelog

See CHANGELOG.md for the full history of changes.

---

Original Claude Code version: taches-cc-resources

—TÂCHES