Potencia tu agente
Instalación del publicador. No es un endpoint de PlugMCP.
Host a copy with our keys and logs.
Aún no hay una versión empaquetada en el registro. Sigue el README del repositorio.
¿URL remota o un comando en tu máquina? En qué se diferencian las dos instalaciones
Ejecutar en PlugMCP
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
AbrirMost RAG tools make you choose between cloud power and data privacy. Axon is local-first — full capability with zero egress when you run on Ollama or vLLM; cloud providers (OpenAI, Gemini, Grok, GitHub Copilot, Ollama Cloud) stay optional.
@axon in Copilot Chat, MCP for Claude Code / Codex / Gemini CLI / Cursor, Graph panel in VS Code or your browser.sources + citations arrays with character offsets (Claude / OpenAI compatible)valid_at + invalid_at--graph-retrieve "..." --graph-at TS--graph-conflicts surfaces status='conflicted' facts@projects, @mounts, @store)Sealed (AES-256-GCM encrypted) sharing works through any cloud sync drive. Files are ciphertext on disk — cloud providers see only encrypted bytes.
normal → draining → readonly → offline/docs@axon VS Code chat participant with Graph panelspip install "axon-rag[starter]" # Python 3.10+. Sealed sharing + extra loaders. Web GUI ships with axon-api.
axon --ingest ./my-notes # index a folder: local embeddings, no LLM calls
axon "What do my notes say about the release plan?" # cited answer from a local Ollama model
The first question pulls the default model (llama3.1:8b) into Ollama if
it is missing, or use a cloud model: axon --model gpt-4o-mini "…" with OPENAI_API_KEY set.
axon on its own opens the REPL; on a fresh machine it runs the setup wizard first.
If something doesn't look right:
axon --doctor # Health checks: Python, Ollama, model pulled, store writable.
axon update # Check PyPI and upgrade the package + VS Code extension together.
Local inference uses Ollama or vLLM (self-hosted). Cloud providers (OpenAI, Gemini, Grok, GitHub Copilot, Ollama Cloud) work via API keys.
Share an encrypted knowledge base through OneDrive, Dropbox, or Google Drive. Cloud providers see only ciphertext.
pip install "axon-rag[starter]" # on both machines (or the smaller [sealed] extra)
Owner
axon --store-init "/path/to/OneDrive/axon" # 1. put the store in the synced folder
axon --project-new research --ingest /path/to/documents # 2. create a project and index it
axon # 3. open the REPL for the sealed steps:
axon> /store bootstrap <passphrase> first time only; later sessions: /store unlock <passphrase>
axon> /project seal research encrypt the project in place
axon> /share generate research alice prints the share string — send it to alice
Grantee
axon --store-init "/path/to/OneDrive/axon" # 1. the same synced folder
axon --share-redeem "<share string>" # 2. key goes to the OS keyring
axon --project mounts/owner_research "question" # 3. decrypted to a temp cache, wiped on exit
→ Full Sharing Guide — OneDrive / Dropbox / Google Drive setup, revocation, expiry, filesystem compatibility matrix.
Browser UI: start axon-api and open http://localhost:8420/gui/. No extra
command or dependency needed — the GUI ships with the server.
Single-instance routing. If an
axon-apiserver is already running, theaxonCLI detects it (a quick/health/readyprobe) and routes store-mutating commands —--ingestand project create / delete / switch — through that server instead of opening a second in-process copy of the same store. This avoids two processes racing on the vector-store files, skips reloading the embedding model on every CLI call, and keeps a single writer for both the store and the logs. Pass--localto force an in-process brain, or point the CLI at a specific server with theAXON_API_BASEenvironment variable (or theapi_host/api_portconfig fields).Relatedly,
axon-apiitself refuses to start a second server on a store anotheraxon-apiis already serving (one writer per store) — setAXON_ALLOW_MULTIPLE_SERVERS=1if you really want two.
Install the bundled VSIX to unlock the @axon chat participant, Knowledge Graph panel and Code Graph panel — directly inside VS Code alongside Copilot.
Extensions panel → "..." → Install from VSIX...
→ run `axon-ext` (or install from VSIX manually)
Or connect via MCP for Copilot agent mode — point .vscode/mcp.json at axon-mcp and all 18 tools appear in the agent hammer menu automatically.
The VS Code extension surfaces 20 LM tools to Copilot Chat — the 18 MCP tools (query, search, ingest, config, graph facts, sharing) plus
show_graphandingest_image.
Drop-in retrievers for LangChain and LlamaIndex agents — no REST round-trips, no extra process. Both wrap the same AxonBrain.search_raw() codepath the REST and REPL surfaces use, so hybrid search, reranking, HyDE, multi-query, and the GraphRAG budget apply automatically.
# pip install "axon-rag[langchain]"
from axon import AxonBrain, AxonConfig
from axon.integrations.langchain import AxonRetriever
brain = AxonBrain(AxonConfig.load()) # ~/.config/axon/config.yaml, or .load("path.yaml")
retriever = AxonRetriever(brain=brain, top_k=5)
docs = retriever.invoke("what does the project do?") # list[Document]
# pip install "axon-rag[llama-index]"
from axon.integrations.llama_index import AxonLlamaRetriever
retriever = AxonLlamaRetriever(brain=brain, top_k=5)
nodes = retriever.retrieve("what does the project do?") # list[NodeWithScore]
Per-call overrides (e.g. force HyDE for one question): retriever.with_overrides({"hyde": True}).invoke(query). From async code, await retriever.aretrieve(query, hyde=True, top_k=8) accepts the same flags as kwargs without rebuilding the retriever (v0.4.1).
Ingestion is sandboxed to a configurable base directory (RAG_INGEST_BASE). Requests outside it are rejected with 403. See SECURITY.md.
MIT — see LICENSE.
Local-first RAG engine with cited answers from your own docs. GraphRAG, sealed sharing, MCP-native, VS Code Copilot integration. Nothing leaves your machine.
Herramientas
Aún no hay lista de herramientas. Los servidores remotos la reciben tras la primera comprobación de estado.
Preguntas frecuentes
Copia en esta página la configuración stdio para Cursor, Claude Desktop, Claude Code o VS Code. Usa la instalación del publicador, no un proxy de PlugMCP.
Este registro no indica ninguna variable de entorno obligatoria.
Axon se ejecuta localmente por stdio.
Última comprobación: 2026-10-28. Local-first RAG engine with cited answers from your own docs. GraphRAG, sealed sharing, MCP-native, VS Code Copilot integration. Nothing leaves your machine.