- Home
- Skills
- Agents & Orchestration
- Stxic Agent Foundry
More screenshots
Works with the AI Agents you already use
Stxic Agent Foundry
Describe the AI teammates you want. Get purpose-built subagent definitions and a practical delegation plan for your AI tool.
$3
Stxic Agent Foundry
Example session with this skill installed
I need a team for my Next.js project: one agent to write Tailwind components from screenshots and another to write Playwright tests for them. Use Cursor format.
- Read your context and instructions
- Compiled the stxic agent foundry
I have designed a two-agent team for your Next.js project. The ui-specialist handles component implementation, while the qa-engineer ensures coverage. I have placed the native definitions in your .cursor/agents/ directory.
Subagent Team Design
| Agent | Trigger | Responsibility | Deliverable |
|---|---|---|---|
ui-specialist | Image provided | Convert screenshots to Tailwind + Lucide components | .tsx component files |
qa-engineer | Component created | Write E2E Playwright tests for new components | .spec.ts test files |
Native Agent Definitions
# .cursor/agents/ui-specialist.md
---
name: ui-specialist
description: Expert in React, Tailwind CSS, and shadcn/ui component generation from visual specs.
agent_mode: subagent
---
You are a specialist UI developer. Your goal is to implement pixel-perfect React components using Tailwind CSS.
- Input: Screenshot or wireframe description.
- Constraints: Use TypeScript, follow existing project styling, and ensure accessibility (ARIA).
- Boundary: Do not write business logic or database queries.
# .cursor/agents/qa-engineer.md
---
name: qa-engineer
description: Playwright specialist for E2E and component testing in Next.js environments.
agent_mode: subagent
---
You are a QA automation engineer. Your goal is to ensure 100% test coverage for new UI components.
- Input: Path to a new React component.
- Responsibilities: Write Playwright tests covering happy paths, edge cases, and interactions.
- Verification: Run tests and report failures before marking a task complete.
Orchestration Plan
- Handoff: Once
ui-specialistwrites a file tocomponents/, it must notify the coordinator. - Sequence: The coordinator then invokes
qa-engineerwith the file path. - Completion: The task is done only when
qa-engineerreports a passing test suite.
Next steps
- Run the smoke-test: "@ui-specialist create a hero section based on /docs/wireframe.png"
- Check
.cursor/agents/to verify the two new files were created. - Update your
.cursorrulesto include these agents in the general workflow.
Connects securely to your tools. The creator never sees your data.
About this skill
Stop wasting time manually writing system prompts and figuring out how to link AI agents together. This skill transforms plain-English project descriptions into focused subagent teams with clear delegation logic.
What it does
- Design specialist teams by converting high-level goals into 1-5 distinct agent roles with specific triggers.
- Generate native configurations for Claude Code, Cursor, Codex, Gemini CLI, OpenCode, and GitHub Copilot.
- Create delegation contracts that define exactly how agents hand off work, share context, and verify results.
- Audit existing agents to find overlapping responsibilities, conflicting instructions, or missing safety boundaries.
- Migrate agent definitions between different AI platforms while maintaining consistent behavior and tool access.
How it works
- Analyze requirements to extract the domain, project environment, and desired autonomy from your request.
- Architect the team by defining the minimum viable roles, sequential vs parallel workflows, and specific deliverables.
- Format for your host by selecting the correct native file structure (YAML, TOML, or Markdown) for your IDE or CLI.
- Install and verify by writing files directly to your project-scoped paths (e.g.,
.cursor/agents/) and providing a smoke-test prompt.
Frameworks & tools
Supported native exports include .claude/agents/, .cursor/agents/, .codex/agents/, .gemini/agents/, .opencode/agents/, and .github/agents/. It also generates portable Markdown cards for generic chat environments.
Why this beats prompting it yourself
Writing subagent prompts requires balancing narrow focus with enough context to be useful. This skill enforces "least privilege" safety, prevents role overlap, and handles the specific boilerplate formatting required by different IDE agent runners so you do not have to look up documentation.
Use cases
- Scaffold a frontend/backend agent pair that coordinates through a shared API contract.
- Turn a messy README into a research team that audits competitors and drafts feature specs.
- Convert your custom Claude prompts into native Cursor subagents with proper YAML frontmatter.
- Review a repository's existing agent folder to fix logic loops or permission overreach.
Known limitations
This skill creates the definitions and instructions but does not grant subagent execution capabilities to AI clients that do not natively support them. It cannot bypass host-level tool permissions.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
Does not add subagent execution to unsupported clients. Cannot bypass host-level tool permissions or OS-level security boundaries. Requires file-write access for automatic installation.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 7 days ago