playbooks

home / skills / jeremylongshore / claude-code-plugins-plus-skills / windsurf-cascade-agents

windsurf-cascade-agents skill

/plugins/saas-packs/skill-databases/windsurf/skills/windsurf-cascade-agents

This skill helps you create and configure customized Cascade agents tailored to domain tasks, enabling targeted prompts and activation.

npx playbooks add skill jeremylongshore/claude-code-plugins-plus-skills --skill windsurf-cascade-agents

Review the files below or copy the command above to add this skill to your agents.

Files (5)
SKILL.md
2.1 KB
---
name: "windsurf-cascade-agents"
description: |
  Create custom Cascade agent configurations for specialized tasks. Activate when users mention
  "custom cascade agent", "specialized ai agent", "domain-specific cascade", "agent configuration",
  or "custom ai behavior". Handles custom agent creation and configuration. Use when working with windsurf cascade agents functionality. Trigger with phrases like "windsurf cascade agents", "windsurf agents", "windsurf".
allowed-tools: "Read,Write,Edit,Bash(cmd:*),Grep,Glob"
version: 1.0.0
license: MIT
author: "Jeremy Longshore <[email protected]>"
---

# Windsurf Cascade Agents

## Overview

This skill enables creation of specialized Cascade agents tailored for specific domains or tasks. Custom agents can be configured with domain knowledge, specialized prompts, and focused capabilities. Use cases include security review agents, API design agents, performance optimization agents, and documentation writers that understand your project's specific conventions and requirements.

## Prerequisites

- Windsurf IDE with Cascade Pro or Enterprise
- Understanding of prompt engineering principles
- Domain knowledge to encode in agent context
- Project documentation and conventions documented
- Test scenarios for agent validation

## Instructions

1. **Define Agent Purpose**
2. **Create System Prompt**
3. **Configure Context Sources**
4. **Set Activation Triggers**
5. **Test and Refine**


See `{baseDir}/references/implementation.md` for detailed implementation guide.

## Output

- Custom agent configurations in registry
- Domain-specific system prompts
- Context files with specialized knowledge
- Activation trigger mappings

## Error Handling

See `{baseDir}/references/errors.md` for comprehensive error handling.

## Examples

See `{baseDir}/references/examples.md` for detailed examples.

## Resources

- [Windsurf Custom Agents Guide](https://docs.windsurf.ai/features/custom-agents)
- [Prompt Engineering Best Practices](https://docs.windsurf.ai/guides/prompt-engineering)
- [Agent Context Management](https://docs.windsurf.ai/features/context-management)

Overview

This skill creates custom Cascade agent configurations tailored to specialized tasks and domains. It helps encode domain knowledge, system prompts, context sources, and activation triggers so agents act with predictable, focused behavior. Use it to produce registry-ready agent configurations and testable agent behaviors for Windsurf Cascade deployments.

How this skill works

You define an agent purpose and craft a domain-aware system prompt, then attach context sources such as docs, code, or examples. The skill maps activation triggers and runtime parameters, generates configuration artifacts, and provides test scenarios so you can iterate until the agent meets validation criteria. Outputs include registry entries, prompt files, context bundles, and trigger mappings ready for Windsurf Cascade.

When to use it

  • You need a domain-specific assistant that understands project conventions or security rules.
  • You want automated reviewers for code, APIs, or performance with custom scoring criteria.
  • You need specialized writing agents for docs, release notes, or compliance content.
  • You want to reduce human tuning by encoding expertise directly into an agent.
  • You are deploying multiple agents with consistent activation and versioning policies.

Best practices

  • Start with a precise agent purpose and measurable success criteria before authoring prompts.
  • Embed concise, high-value context documents rather than large, unfocused corpora.
  • Create clear activation triggers and fallbacks to avoid accidental or inappropriate activations.
  • Include validation scenarios and unit-style tests for each agent behavior change.
  • Version prompts and configuration files so you can roll back or A/B test changes.

Example use cases

  • Security review agent that flags insecure patterns and suggests fixes using project-specific rules.
  • API design agent that enforces API style guides and generates OpenAPI examples from code.
  • Performance optimization agent that reviews profiling output and recommends hotspot fixes.
  • Documentation writer that knows project terminology and produces release notes and component docs.
  • DevOps agent that validates CI/CD manifests against organizational policies.

FAQ

What inputs does the skill require to create an agent?

Provide the agent purpose, a system prompt, prioritized context sources (docs, code snippets, examples), activation trigger phrases, and validation scenarios.

How do I validate an agent before production?

Use the included test scenarios to run sample conversations, check outputs against success criteria, and iterate on prompts and context until results meet your metrics.