knowledgexlab/heta/querying-knowledge-and-memory

Claude Code and agent skill by knowledgexlab/heta

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Use when the agent needs to answer questions from stored documents, recall past observations or context, store new knowledge for later, or discover available knowledge bases. Triggers on any of: "search the knowledge base", "what do you know about X", "remember this", "recall what happened", "find in documents", "store this observation", "look it up", "do you remember". Covers HetaDB (human document KBs), MemoryKB (agent knowledge graph), and MemoryVG (personal episodic memory).

Install

npx vskill@latest install knowledgexlab/heta/querying-knowledge-and-memory

SKILL.md

Querying Knowledge and Memory

Three Systems — Know Which to Use

SystemBuilt byRole
HetaDBHumans (file upload)Multimodal document knowledge bases
MemoryKBAgent (text insert)Agent's own accumulating knowledge graph
MemoryVGAgent (conversation)Fast personal memory & cross-session cache

Base URL: http://<host>:8000


Orchestration — Always Follow This Order

When answering a knowledge question:

Step 1 — Search MemoryVG (fast personal recall)
→ Answer is clear and specific → return it, done
→ Vague or absent → continue to Step 2
Step 2 — Query HetaDB (human knowledge base)
→ Synthesised answer + citations
Step 3 — Agent decides what to store:
→ Quick recall next time? → MemoryVG add
→ Worth accumulating? → MemoryKB insert
→ Ephemeral / trivial? → skip

API Summary

HetaDB — Document Knowledge Bases
GET /api/v1/hetadb/files/knowledge-bases

→ Returns data[].name — use as kb_id. Skip KBs with status: "deleting"; all others are queryable.

POST /api/v1/hetadb/chat
{ "query": "...", "kb_id": "...", "user_id": "agent", "query_mode": "naive" }

→ Required: query, kb_id. user_id is always "agent". Read response field from result. → Full query mode guide: see hetadb-api.md


MemoryVG — Personal Memory Cache
POST /api/v1/hetamem/vg/search
{ "query": "...", "agent_id": "agent" }

→ agent_id is always "agent". Read results[].memory.

POST /api/v1/hetamem/vg/add
{ "messages": [{"role": "assistant", "content": "..."}], "agent_id": "agent" }

→ agent_id is always "agent". LLM extracts facts from messages. Available immediately.


MemoryKB — Agent Knowledge Graph
POST /api/v1/hetamem/kb/insert (multipart/form-data)
query=<text to add to knowledge graph>

→ Async — LightRAG graph construction takes ~200s. Do not query immediately.

POST /api/v1/hetamem/kb/query
{ "query": "...", "mode": "hybrid" }

→ Read final_answer from result.


When NOT to Use Each

Don't useFor
HetaDBStoring agent observations (no insert API)
MemoryKBQuick recall — async delay makes it unsuitable as cache
MemoryVGDeep multi-document reasoning — use HetaDB instead

Full parameter reference: hetadb-api.md, hetamem-api.md

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
14
Repository forks
1
7-day trend score
19

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/knowledgexlab/heta/querying-knowledge-and-memory/badge)](https://verified-skill.com/skills/knowledgexlab/heta/querying-knowledge-and-memory)

Something wrong with this skill or its page? Tell us.