docs.terraphim.ai cleanup: Add graph embeddings, quickstart guides, and OpenClaw/KimiClaw integration
@AlexMikhalev 已经在做这个了。
开始于 2026年3月10日。
评估
这个 Issue 还没有评估数据。
描述
Proposal: Documentation Cleanup for docs.terraphim.ai
Background
The current documentation at docs.terraphim.ai needs restructuring to better serve new users and showcase Terraphim's unique capabilities. This issue proposes adding several key sections to improve onboarding and highlight the quantum-inspired knowledge graph architecture.
Proposed Additions
1. Why Graph Embeddings? 🧠
Purpose: Explain the core philosophy behind Terraphim's knowledge graph approach
Content to include:
- What are graph embeddings and why they matter for AI
- Comparison: Vector-only RAG vs. Graph-enhanced RAG
- Terraphim's quantum-inspired graph representation
- Real-world benefits: multi-hop reasoning, explainability, relationship preservation
- Performance benchmarks (if available)
Target audience: Technical decision-makers, AI engineers evaluating knowledge systems
2. Quickstart Guide ⚡
Purpose: Get new users up and running in under 5 minutes
Content to include:
- Installation (cargo install, pre-built binaries, Docker)
- Minimal working example: indexing a document
- Querying the knowledge graph
- Basic configuration
- Verification steps
Structure:
# Step 1: Install
$ cargo install terraphim
# Step 2: Index
$ terraphim index ./my-docs
# Step 3: Query
$ terraphim query "What is the main topic?"
3. Quickstart with OpenClaw 🐾
Purpose: Integration guide for OpenClaw users
Content to include:
- What is OpenClaw and why integrate with Terraphim
- Setting up the Terraphim skill for OpenClaw
- Configuration: connecting OpenClaw to local/remote Terraphim instance
- Example: Agent memory persistence via knowledge graph
- Example: Multi-agent coordination using shared graph state
- Troubleshooting common issues
Key integration points:
- OpenClaw's skill system → Terraphim knowledge graph
- Agent session memory → Persistent graph storage
- Cross-agent communication → Meta-cortex formation
4. Quickstart with KimiClaw 🤖
Purpose: Guide for KimiClaw (Kimi K2.5 + OpenClaw) users
Content to include:
- What is KimiClaw (Kimi K2.5 model + OpenClaw runtime)
- Why Kimi K2.5 pairs well with Terraphim's knowledge graph
- Setup instructions specific to KimiClaw environment
- Example: Long-context conversations with graph-backed memory
- Example: Code understanding with semantic code graph
- Performance tips for the K2.5 + Terraphim stack
Unique advantages:
- Kimi K2.5's 256K context + Terraphim graph = best of both worlds
- Structured knowledge extraction from long conversations
- Persistent memory across sessions
Suggested Documentation Structure
docs.terraphim.ai/
├── index.md # Landing page
├── why-graph-embeddings.md # NEW: Philosophy & benefits
├── quickstart/
│ ├── index.md # NEW: Basic quickstart
│ ├── openclaw.md # NEW: OpenClaw integration
│ └── kimiclaw.md # NEW: KimiClaw integration
├── concepts/
│ ├── knowledge-graph.md
│ ├── quantum-fields.md
│ └── meta-cortex.md
└── ...
Additional Considerations
- Cross-references: Link between sections (e.g., OpenClaw guide references "Why Graph Embeddings")
- Code examples: All commands should be copy-paste runnable
- Screenshots: Diagrams for graph structure, meta-cortex formation
- Video: Optional short screencasts for complex setups
Related Work
This documentation effort aligns with:
- Terraphim's positioning as "knowledge-centric AI"
- Recent marketing story emphasizing quantum-inspired architecture
- Growing OpenClaw ecosystem integration
Next Steps
- Review and approve scope
- Assign documentation contributors
- Create sub-issues for each section
- Set review milestones
Would love feedback on:
- Priority ordering of these sections
- Additional integration guides (e.g., other agent frameworks)
- Specific use cases to highlight in examples
cc: @AlexMikhalev (KimiClaw perspective)
- 主要语言
- Rust
- 星标
- 66
- 派生
- 5
- 平均合并
- 1 小时
- 30 天内合并 PR
- 3
环境准备
- 没有 Dockerfile 或 Docker Compose 文件
- 有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
terraphim/terraphim-ai 的其他 Issue
-
terraphim_rlm: require a per-sub-result verification plan (recheck / control / independent route)未关闭
难度 4/5 3-5 天 新手友好度 48/100
terraphim/terraphim-ai#969 · 1 条评论 ·
-
难度 5/5 一周以上 新手友好度 35/100
terraphim/terraphim-ai#968 · 1 条评论 ·
-
难度 5/5 一周以上 新手友好度 35/100
terraphim/terraphim-ai#967 ·
-
难度 4/5 3-5 天 新手友好度 48/100
terraphim/terraphim-ai#966 ·
-
难度 5/5 一周以上 新手友好度 30/100
terraphim/terraphim-ai#965 · 2 条评论 ·
查看 terraphim/terraphim-ai 的全部 Issue
相似的 Issue
-
proxy: SQL-over-HTTP turns empty arrays into [""] or [false], or HTTP 500 for int and float arrays未关闭
难度 2/5 1-3 小时 新手友好度 83/100
neondatabase/neon#12965 ·
-
难度 2/5 1-3 小时 新手友好度 78/100
goauthentik/authentik#26867 ·
维护者通常 1 天内回复
-
api bug macos maintainer-needed p2 triaged
难度 2/5 1-3 小时 新手友好度 66/100
维护者通常 1 天内回复
-
area::tern difficulty::easy good first issue kind::cleanup triage::ready
难度 1/5 1-3 小时 新手友好度 88/100
-
难度 2/5 1-3 小时 新手友好度 76/100
eclipse-zenoh/zenoh#2842 · 2 条评论 ·
维护者通常 1 天内回复