Setup
Set up Fiber AI tools in the current environment. Use when the user needs to configure API keys, verify MCP connection, or install SDKs.From its SKILL.md
npx -y skills add fiber-ai/fiber-ai-plugin --skill setupAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
3.3 KB, as published. Nobody here has run it
Fiber AI Setup
Configure Fiber AI for the current environment.
When to Use
- User says "setup fiber", "install fiber", "configure fiber"
- MCP tools are not responding or returning authentication errors
- User is missing an API key or needs to verify their configuration
- First time using Fiber AI in this environment
Setup Steps
1. Check API Key
Check if the FIBER_API_KEY environment variable is set in the current shell.
- If set: verify it works by calling the credits endpoint via Core MCP:
call_operation("getOrgCredits", {"query": {"apiKey": "..."}})If the call succeeds, show "Fiber AI is connected" with the credit balance. - If not set: instruct the user to:
- Get their API key from https://fiber.ai/app/api
- Set it as an environment variable:
export FIBER_API_KEY=sk_live_... - For persistence, add the export to their shell profile (
~/.zshrcor~/.bashrc)
2. Verify MCP Connection
After the API key is configured:
- Try calling a Core MCP tool like
list_all_endpointsto verify connectivity - If successful: confirm "Fiber AI MCP is connected and working."
- If unsuccessful: check that the editor supports HTTP (Streamable HTTP) MCP transport and that network access to
mcp.fiber.aiis not blocked
3. SDK Installation (Optional)
Only if the user intends to write application code (not just use MCP tools):
- TypeScript:
npm install @fiberai/sdk - Python:
pip install fiberai
4. Quick Verification
Suggest the user run a quick test to confirm everything works:
Try running
/fiber:search "technology companies in San Francisco"to verify your setup.
Authentication Notes
- For POST/PATCH/PUT API endpoints:
apiKeyis passed in the request body - For GET API endpoints:
apiKeyis passed in the query string - Check https://api.fiber.ai/docs/ for the exact authentication format per endpoint
Troubleshooting
- MCP not connecting: verify the editor supports HTTP (Streamable HTTP) MCP transport
- Authentication errors (401): API key is invalid or expired — regenerate at https://fiber.ai/app/api
- Insufficient credits (402): top up at https://www.fiber.ai/app/subscription
- Rate limit errors (429): wait a moment and retry
- Network errors: ensure outbound HTTPS to
mcp.fiber.aiis allowed by firewall or proxy
For AI agents: machine-readable docs
- Start here: https://api.fiber.ai/llms.txt — routing policy + critical rules. Read this before making any API call.
- Operation index: https://api.fiber.ai/ai-docs/index.md — every public operation, grouped by tag.
- Per-operation pages: https://api.fiber.ai/ai-docs/{operationId}.md. Drop this straight into the LLM prompt as the canonical context for the operation you're about to call.
- Content negotiation: send
Accept: text/markdownto https://api.fiber.ai/openapi.json to get the agent-friendly index at the same URL (Stripe pattern). - MCP: https://mcp.fiber.ai/mcp/v2 (API key) or https://mcp.fiber.ai/mcp/v3 (OAuth via Clerk).
- MCP quickstart: https://docs.fiber.ai/article/using-mcp-in-llms.