# QevosAgent - AI Agent Friendly Website # Last updated: 2026-05-18 ## Website Overview QevosAgent is a free, open-source local AI Agent supporting Qwen3, DeepSeek V4, Gemma 4, Claude and more models. Features: persistent memory, tool calling, web dashboard, multi-model support. ## Main Pages - Home: https://qevos.ai/ (English) / https://qevos.ai/cn.html (Chinese) - Blog: https://qevos.ai/blog_en.html (English) / https://qevos.ai/blog_cn.html (Chinese) - Quick Start: https://qevos.ai/quickstart_en.html - GitHub: https://github.com/HongyunQiu/QevosAgent - Discord: https://discord.gg/tX2BrQmeD7 ## Agent Discovery - /agent-manifest.json - Structured site information (start here) - /agent.md - Agent usage guide - /llms.txt - This file ## Agent API Endpoints (Static JSON) ### Blog Index - GET /api/agent/blog-index.json - Full blog index with all posts metadata - Returns: total_posts, posts array with slug, title, date, summary, tags, content_url, html_url - Cross-language fields (title_zh, summary_zh) included when available ### Search Index (Client-Side Filtering) - GET /api/agent/search.json - Static search index for client-side filtering - Returns: all posts with a 'search_index' field containing normalized text for keyword matching - **Note**: This is a STATIC index. Query parameters are NOT supported. - To search: download the index, then filter client-side by checking if your keyword appears in the 'search_index' field - Example: `posts.filter(p => p.search_index.includes('qwen'))` ### Article Content - GET /api/agent/content/{slug}.json - Full article in structured JSON format - Returns: title, date, summary, tags, sections (parsed from Markdown ## headings) - Each section has: id, heading, content - Chinese sections (sections_zh) included when available - 404 returned for invalid slugs ## Content Structure ### Blog Post Content JSON Schema ``` { "endpoint": "content", "slug": "2025-04-30-introducing-qevosagent", "title": "Article Title", "date": "2025-04-30", "summary": "Brief description", "tags": ["tag1", "tag2"], "language": "en", "html_url": "/blog/en/slug.html", "sections": [ {"id": "section-id", "heading": "Section Title", "content": "Section text..."} ], "title_zh": "中文标题", "summary_zh": "中文摘要", "sections_zh": [...] // Chinese sections when available } ``` ## Usage Guide for Agents 1. Start with /agent-manifest.json to understand the site structure 2. Use /api/agent/blog-index.json to browse all posts 3. Use /api/agent/search.json for client-side keyword search 4. Use /api/agent/content/{slug}.json to get full structured article with sections 5. Avoid parsing HTML directly - use JSON endpoints for efficiency 6. Respect rate limits: 100 requests/minute ## Rate Limits - 100 requests per minute for Agent access - 1000 requests per hour - Please identify yourself with User-Agent header ## Error Handling - 404: Resource not found (invalid slug or URL) - 429: Rate limit exceeded - 500: Server error ## Contact - GitHub Issues: https://github.com/HongyunQiu/QevosAgent/issues - Discord: https://discord.gg/tX2BrQmeD7