[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-5c77e161-4b39-4c4d-b7b4-6dcee44079f9":3,"$f6Ngx88902uaFuOnacBO9d5uRKNmeY6tlKkH1jWvtyLc":42},{"id":4,"title":5,"description":6,"categoryId":7,"moduleId":8,"tags":9,"prompt":10,"icon":11,"source":12,"sourceUrl":13,"authorId":14,"authorName":15,"isPublic":16,"stars":17,"runs":18,"createdAt":19,"updatedAt":19,"module":20,"category":27,"packages":33},"5c77e161-4b39-4c4d-b7b4-6dcee44079f9","recallmax","免费——AI代理的上帝级长文本记忆。注入50万-100万清洁令牌，自动总结并保留语气\u002F意图，将14轮历史压缩至800个令牌。","cat_coding_backend","mod_coding","sickn33,coding","---\nname: recallmax\ndescription: \"FREE — God-tier long-context memory for AI agents. Injects 500K-1M clean tokens, auto-summarizes with tone\u002Fintent preservation, compresses 14-turn history into 800 tokens.\"\ncategory: memory\nrisk: safe\nsource: community\ndate_added: \"2026-03-13\"\nauthor: christopherlhammer11-ai\ntags: [memory, context, rag, summarization, compression, long-context, agent-infrastructure]\ntools: [claude, cursor, codex, gemini, copilot, windsurf, antigravity, grok]\n---\n\n# RecallMax — God-Tier Long-Context Memory\n\n## Overview\n\nRecallMax enhances AI agent memory capabilities dramatically. Inject 500K to 1M clean tokens of external context without hallucination drift. Auto-summarize conversations while preserving tone, sarcasm, and intent. Compress multi-turn histories into high-density token sequences.\n\nFree forever. Built by the Genesis Agent Marketplace.\n\n## Install\n\n```bash\nnpx skills add christopherlhammer11-ai\u002Frecallmax\n```\n\n## When to Use This Skill\n\n- Use when your agent loses context in long conversations (50+ turns)\n- Use when injecting large RAG\u002Fexternal documents into agent context\n- Use when you need to compress conversation history without losing meaning\n- Use when fact-checking claims across a long thread\n- Use for any agent that needs to remember everything\n\n## How It Works\n\n### Step 1: Context Injection\n\nRecallMax cleanly injects external context (documents, RAG results, prior conversations) into the agent's working memory. Unlike naive concatenation, it:\n- Deduplicates overlapping content\n- Preserves source attribution\n- Prevents hallucination drift from context pollution\n\n### Step 2: Adaptive Summarization\n\nAs conversations grow, RecallMax automatically summarizes older turns while preserving:\n- **Tone** — sarcasm, formality, urgency\n- **Intent** — what the user actually wants vs. what they said\n- **Key facts** — numbers, names, decisions, commitments\n- **Emotional register** — frustration, excitement, confusion\n\n### Step 3: History Compression\n\nCompress a 14-turn conversation history into ~800 high-density tokens that retain full semantic meaning. The compressed output can be re-expanded if needed.\n\n### Step 4: Fact Verification\n\nBuilt-in cross-reference checks for controversial or ambiguous claims within the conversation context. Flags contradictions and unsupported assertions.\n\n## Best Practices\n\n- ✅ Use RecallMax at the start of long-running agent sessions\n- ✅ Enable auto-summarization for conversations beyond 20 turns\n- ✅ Use compression before hitting context window limits\n- ✅ Let the fact verifier run on high-stakes outputs\n- ❌ Don't inject unvetted external content without dedup\n- ❌ Don't skip summarization and rely on raw truncation\n\n## Related Skills\n\n- `@tool-use-guardian` - Tool-call reliability wrapper (also free from Genesis Marketplace)\n\n## Links\n\n- **Repo:** https:\u002F\u002Fgithub.com\u002Fchristopherlhammer11-ai\u002Frecallmax\n- **Marketplace:** https:\u002F\u002Fgenesis-node-api.vercel.app\n- **Browse skills:** https:\u002F\u002Fgenesis-marketplace.vercel.app\n\n## Limitations\n- Use this skill only when the task clearly matches the scope described above.\n- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.\n- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.\n","","imported","https:\u002F\u002Fgithub.com\u002Fsickn33\u002Fantigravity-awesome-skills","user_system_seed","SkillOPIC",true,51,1944,"2026-05-16 13:36:43",{"id":8,"name":21,"slug":22,"icon":23,"description":24,"sort":25,"createdAt":26},"编程开发","coding","mdi-code-braces","代码生成、调试、审查，提升开发效率",2,"2026-05-16 12:53:40",{"id":7,"name":28,"slug":29,"icon":30,"description":31,"moduleId":8,"sort":25,"skillCount":32,"createdAt":26},"后端开发","backend","mdi-server","API、数据库、服务端架构",296,[34],{"id":35,"skillId":4,"version":36,"fileName":37,"fileSize":38,"filePath":39,"fileHash":40,"manifest":41,"createdAt":19},"08864054-dd25-42f6-a110-4719b3716cdd","1.0.0","recallmax.zip",1741,"uploads\u002Fskills\u002F5c77e161-4b39-4c4d-b7b4-6dcee44079f9\u002Frecallmax.zip","165f13935e5693e4acec6d6d39863a43a2cb53165287b3ca152bade8c2a64dd3","[{\"path\":\"SKILL.md\",\"isDirectory\":false,\"size\":3362}]",{"code":43,"message":44,"data":45},200,"success",{"items":46,"stats":47,"page":50},[],{"averageRating":48,"totalRatings":48,"ratingCounts":49},0,[48,48,48,48,48],{"limit":51,"offset":48,"hasMore":52,"nextOffset":51,"ratedOnly":16},15,false]