Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Windsurf Sdk Patterns

ASecurity

'Analyze and apply production-ready Devin Desktop (formerly Windsurf) workspace configuration and Cascade

2,785 stars
0 votes
0 copies
0 views
Added 9/24/2026
ai-agentstypescriptgobashsqlnodetestinggitapidatabasesecurity

Works with

claude codeterminalapimcp

Security Analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned 9/24/2026

$npx -y skills add jeremylongshore/claude-code-plugins-plus --skill windsurf-sdk-patterns --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Windsurf Sdk Patterns?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Windsurf Sdk Patterns
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/jeremylongshore-windsurf-sdk-patterns-1cfa7f4b/badge)](https://www.skillsdirectory.com/skills/jeremylongshore-windsurf-sdk-patterns-1cfa7f4b)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: windsurf-sdk-patterns
description: 'Analyze and apply production-ready Devin Desktop (formerly Windsurf) workspace configuration and Cascade
  interaction patterns.

  Use when configuring .devin/rules/project.md, workspace rules, MCP servers,

  or establishing team coding standards for Windsurf AI.

  Trigger with phrases like "windsurf patterns", "windsurf best practices",

  "windsurf config patterns", "windsurfrules", "windsurf workspace".

  '
allowed-tools: Read, Write, Edit, Grep
argument-hint: "[scope or requirements]"
version: 1.12.0
license: MIT
author: Jeremy Longshore <jeremy@intentsolutions.io>
tags:
- saas
- windsurf
- configuration
- rules
- mcp
compatibility: Designed for Claude Code
---
# Windsurf Configuration Patterns

## Overview

Production-ready configuration patterns for Windsurf IDE: rules files, workspace rules with trigger modes, MCP server integration, and Cascade prompt engineering.

## Prerequisites

- Windsurf authenticated and operational
- Understanding of Cascade Code and Chat modes
- Project with established coding conventions

## Tool Use

- Use `Read` to inspect only the repository files and configuration needed for the request.
- Use `Grep` to locate relevant settings, rules, logs, or code without broad collection.
- Use `Write` only for a new artifact the user requested; never write credentials or unreviewed production configuration.
- Use `Edit` for bounded, reviewable changes and preserve unrelated user work.

## Instructions

### Step 1: Root-Level .devin/rules/project.md (Permanent Context)

The `.devin/rules/project.md` file is the single highest-impact configuration for Cascade output quality. It provides persistent context every session.

```markdown
<!-- .devin/rules/project.md -->

# Project: payments-api

## Stack
- Runtime: Node.js 20 LTS
- Language: TypeScript 5.x (strict, noUncheckedIndexedAccess)
- Framework: Fastify v4
- ORM: Drizzle (PostgreSQL)
- Validation: zod
- Testing: Vitest
- Linting: Biome

## Architecture Rules
- Route handlers in src/routes/ — no business logic
- Business logic in src/services/ — never throw, use Result<T,E>
- Database queries in src/repositories/ — Drizzle only
- Shared types in src/types/ — all exported with JSDoc

## Don't
- Don't use `any` type
- Don't use default exports
- Don't use class-based patterns (use functions + closures)
- Don't modify files in migrations/ without explicit request
- Don't use deprecated APIs: my_old_helper, legacyAuth

## Testing
- Unit tests for every service method
- Integration tests for every route handler
- No mocking repositories in integration tests
- Use test fixtures from tests/fixtures/
```

**Limits:** 6,000 characters per rules file. 12,000 total (global + workspace combined).

### Step 2: Workspace Rules with Trigger Modes

Create granular rules in `.devin/rules/` with YAML frontmatter:

```markdown
<!-- .devin/rules/testing.md -->
---
trigger: glob
globs: **/*.test.ts, **/*.spec.ts
---
All test files must:
- Use describe/it blocks (not test())
- Mock external API calls with msw
- Assert both success and error paths
- Include at least one snapshot test for UI components
- Use factory functions from tests/fixtures/ for test data
```

```markdown
<!-- .devin/rules/api-routes.md -->
---
trigger: glob
globs: src/routes/**/*.ts
---
API route handlers must:
- Validate input with zod schema before processing
- Return consistent error format: { error: string, code: string, statusCode: number }
- Include request ID in all log lines
- Never call database directly — use repository layer
```

```markdown
<!-- .devin/rules/security.md -->
---
trigger: model_decision
description: Apply when code touches authentication, authorization, or secrets
---
Security requirements:
- Never log secrets, tokens, or PII
- Use parameterized queries (never string interpolation for SQL)
- Validate JWT tokens with jose library
- Rate limit all public endpoints
- CORS: explicit origin whitelist, never wildcard in production
```

```markdown
<!-- .devin/rules/migrations.md -->
---
trigger: manual
---
Database migration rules (activate with @migrations):
- Always create reversible migrations (up + down)
- Never drop columns in production — deprecate first
- Add indexes for any new foreign key columns
- Test migration on a copy of production data first
```

### Step 3: MCP Server Configuration

Connect external tools to Cascade via Model Context Protocol:

```json
// ~/.codeium/windsurf/mcp_config.json
{
  "mcpServers": {
    "github": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github"],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "${GITHUB_TOKEN}"
      }
    },
    "postgres": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-postgres"],
      "env": {
        "DATABASE_URL": "${DATABASE_URL}"
      }
    },
    "memory": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-memory"]
    }
  }
}
```

Enable in Windsurf Settings > Cascade > Model Context Protocol (MCP).

**Tool limit:** Cascade supports max 100 MCP tools total across all servers. Disable unused tools in each MCP's settings page.

### Step 4: Effective Cascade Prompt Patterns

```
GOOD prompts (specific, scoped):
"In src/services/payment.ts, add a refundPayment method that calls
Stripe's refund API. Handle partial refunds. Return Result<Refund, PaymentError>."

"@src/routes/users.ts Add input validation using the UserCreateSchema
from src/types/user.ts. Return 400 with field-level errors."

BAD prompts (vague, unscoped):
"Add validation to the API"
"Refactor the codebase"
"Make it better"
```

## Output

Create a minimal customization set using the correct mechanism—Rule, `AGENTS.md`, Workflow, Skill, Hook, or MCP configuration—with activation behavior, ownership, secret handling, and a verification example. Avoid undocumented settings keys.

## Error Handling

| Issue | Cause | Solution |
|-------|-------|----------|
| Rules ignored by Cascade | File over 6,000 chars | Trim to essentials, split into workspace rules |
| Workspace rules not loading | Wrong directory or invalid frontmatter | Use `.devin/rules/*.md` and validate the `trigger:` mode |
| MCP server not connecting | Command not found | Ensure npx can resolve the package |
| Too many MCP tools | Over 100 tool limit | Disable unused tools per MCP server |
| Glob trigger not firing | Wrong pattern syntax | Use gitignore-style globs: `**/*.test.ts` |

## Examples

### Global Rules (Apply to All Projects)

```markdown
<!-- ~/.codeium/windsurf/memories/global_rules.md (6,000 char limit) -->
- Always use English for code comments and commit messages
- Prefer functional programming patterns over OOP
- Write self-documenting code; add comments only for "why", not "what"
- When suggesting terminal commands, explain what they do
- Never suggest installing global npm packages
```

### Project Health Check

```bash
# Verify Windsurf config exists
ls -la .devin/rules/project.md .codeiumignore .devin/rules/ 2>/dev/null
```

## Resources

- [Focused first-party references](references/official-docs.md)
- [Windsurf Rules Directory](https://windsurf.com/editor/directory)
- [Cascade Memories](https://docs.devin.ai/desktop/cascade/memories)
- [MCP Integration](https://docs.devin.ai/desktop/cascade/mcp)

## Related Skill

Continue with `windsurf-core-workflow-a` to apply these configuration patterns in a bounded Cascade session with explicit validation and rollback checkpoints.

Attribution

jeremylongshorejeremylongshore
View sourceSee grades on GitHubMore from jeremylongshore →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698621 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →