Skip to main content 홈 크리에이터 daydreamsai skills-market lucid-client-api
lucid-client-api Provides Lucid Client API workflows for the multi-agent runtime.
Covers endpoints for managing agents, invoking entrypoints, and handling
payments in the lucid-client codebase.
Use when interacting with Lucid Client API endpoints, managing agents, or
working with the multi-agent runtime system.
설치로 이동 Skills Marketplace 커뮤니티가 만든 AI 스킬을 발견하고 탐색하세요.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/daydreamsai/skills-market --skill lucid-client-api명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
Zip 다운로드 다운로드 중... name lucid-client-api description Provides Lucid Client API workflows for the multi-agent runtime.
Covers endpoints for managing agents, invoking entrypoints, and handling
payments in the lucid-client codebase.
Use when interacting with Lucid Client API endpoints, managing agents, or
working with the multi-agent runtime system.
see-also [{"https://github.com/daydreamsai/lucid-client/blob/master/AGENTS.md":"Full documentation of the lucid-client architecture"},{"https://github.com/daydreamsai/lucid-client/blob/master/CLAUDE.md":"Development guide for lucid-client"}]
Lucid Client API Skill
This skill provides capabilities to interact with the Lucid Client API, which is a multi-agent runtime system hosted in the lucid-client codebase.
API Base URL
The API base URL is typically:
Development : http://localhost:8787Production : Configured via VITE_API_URL environment variable
Authentication
Most endpoints require authentication via Better Auth:
Session cookies are automatically sent with requests
Protected routes: /api/agents/*, /api/secrets/*
Public routes: /agents/{slug}, /agents/{agentId}/.well-known/agent-card.json
Key Endpoints
Agent Management
List Agents
GET /api/agents
Query params: ?page=1&limit=10&search=query&enabled=true
Returns paginated list of agents for the authenticated owner.
Create Agent
POST /api/agents
Content-Type: application/json
{
"slug": "my-agent",
"name": "My Agent",
"description": "Agent description",
"version": "1.0.0",
"entrypoints": [...],
"paymentsConfig": {...},
"walletsConfig": {...}
}
Get Agent GET /api/agents/{agentId}
Update Agent PUT /api/agents/{agentId}
Content-Type: application/json
{
"name": "Updated Name",
"description": "Updated description",
...
}
Delete Agent DELETE /api/agents/{agentId}
Get Public Agent by Slug Public endpoint (no auth required). Returns enabled agents only.
Agent Invocation
Get Agent Manifest GET /agents/{agentId}/.well-known/agent-card.json
Returns A2A-compatible agent manifest/card.
List Entrypoints GET /agents/{agentId}/entrypoints
Returns all entrypoints for an agent.
Invoke Entrypoint POST /agents/{agentId}/entrypoints/{key}/invoke
Content-Type: application/json
{
"input": {
"field1": "value1",
"field2": "value2"
},
"metadata": {
"requestId": "optional-request-id"
}
}
If entrypoint requires payment, returns 402 Payment Required
Response includes x402 payment headers:
X-Payment-Required: trueX-Payment-Price: "$0.001"X-Payment-Network: "eip155:84532"X-Payment-PayTo: "0x..."
Client must complete payment and retry with payment headers:
PAYMENT-REQUIRED: Payment amountPAYMENT-SIGNATURE: Payment signaturePAYMENT-RESPONSE: Payment response
Secrets Management
List Secrets GET /api/agents/{agentId}/secrets
Create Secret POST /api/agents/{agentId}/secrets
Content-Type: application/json
{
"key": "API_KEY",
"value": "secret-value"
}
Get Secret GET /api/agents/{agentId}/secrets/{key}
Update Secret PUT /api/agents/{agentId}/secrets/{key}
Content-Type: application/json
{
"value": "new-secret-value"
}
Delete Secret DELETE /api/agents/{agentId}/secrets/{key}
Analytics
Get Analytics Summary GET /api/agents/{agentId}/analytics/summary
Query params: ?startDate=2024-01-01&endDate=2024-01-31
Get Analytics Transactions GET /api/agents/{agentId}/analytics/transactions
Query params: ?page=1&limit=10&startDate=...&endDate=...
Invocations
List Invocations GET /api/agents/{agentId}/invocations
Query params: ?page=1&limit=10&status=success&entrypointKey=echo
Get Invocation GET /api/agents/{agentId}/invocations/{invocationId}
Get Invocation Stats GET /api/agents/{agentId}/invocations/stats
Query params: ?startDate=...&endDate=...
Identity
Retry Identity Registration POST /api/agents/{agentId}/identity/retry
Update Identity Metadata PUT /api/agents/{agentId}/identity/metadata
Content-Type: application/json
{
"metadata": {
"key": "value"
}
}
Rankings
Get Rankings GET /api/rankings
Query params: ?window=24h&limit=10
Stream Rankings GET /api/rankings/stream
Query params: ?window=24h
Server-Sent Events stream of live rankings.
Health & Stats
Health Check
Redis Health
Network Stats
OpenAPI Documentation The API provides full OpenAPI documentation:
OpenAPI Spec : GET /doc (JSON)Swagger UI : Available when served (typically at /swagger)
Common Patterns
Making Authenticated Requests When making requests from code, include credentials:
const response = await fetch (`${API_URL} /api/agents` , {
method : 'GET' ,
credentials : 'include' ,
headers : {
'Content-Type' : 'application/json' ,
},
});
Handling Payment-Required Responses let response = await fetch (`${API_URL} /agents/${agentId} /entrypoints/${key} /invoke` , {
method : 'POST' ,
credentials : 'include' ,
headers : {
'Content-Type' : 'application/json' ,
},
body : JSON .stringify ({ input : {...} }),
});
if (response.status === 402 ) {
const price = response.headers .get ('X-Payment-Price' );
const network = response.headers .get ('X-Payment-Network' );
const payTo = response.headers .get ('X-Payment-PayTo' );
const paymentResponse = await completePayment ({ price, network, payTo });
response = await fetch (`${API_URL} /agents/${agentId} /entrypoints/${key} /invoke` , {
method : 'POST' ,
credentials : 'include' ,
headers : {
'Content-Type' : 'application/json' ,
'PAYMENT-REQUIRED' : price,
'PAYMENT-SIGNATURE' : paymentResponse.signature ,
'PAYMENT-RESPONSE' : paymentResponse.response ,
},
body : JSON .stringify ({ input : {...} }),
});
}
Error Handling The API returns standard error responses:
{
"error" : "Error type" ,
"message" : "Human-readable error message" ,
"details" : {...}
}
400: Validation error401: Unauthorized (not authenticated)402: Payment required404: Resource not found409: Conflict (e.g., slug already exists)500: Internal server error
When to Use This Skill
User wants to interact with the Lucid Client API
User needs to manage agents programmatically
User wants to invoke agent entrypoints
User needs to handle payments for agent invocations
User wants to query analytics or rankings
User needs to manage agent secrets
Notes
The API is in the lucid-client codebase, not lucid-agents
The API uses Hono with OpenAPI validation
All agent operations are scoped to the authenticated owner
Payments use the x402 protocol (HTTP-native payments)
The runtime is stateless - agents are built dynamically from stored definitions
