AllocContext
Self-host MCP: portfolio crypto context for agents. PyPI stdio; MIT.
Install
Add this to your client's MCP configuration (Claude Desktop: claude_desktop_config.json; Cursor: .cursor/mcp.json; VS Code and others use the same "mcpServers" shape).
alloc-contextclaude mcp add alloccontext -- uvx alloc-context==0.3.0 mcp
{
"mcpServers": {
"alloccontext": {
"command": "uvx",
"args": [
"alloc-context==0.3.0",
"mcp"
]
}
}
}
Documentation
From the project's README on GitHub, refreshed weekly. Links open on GitHub.
Contents
AllocContext
mcp-name: io.github.AllocContext/alloc-context
Portfolio-aware crypto context for whatever you hold — discover your holdings, holdings-scoped market data, sentiment, macro, and regime; optional allocation analysis when you supply targets. Deterministic JSON over MCP.
New here? Cursor MCP setup — stdio in your editor, or self-hosting with local ingest. Organization: AllocContext on GitHub.
Privacy: nothing stored · one-time read-only · pass-through only when using live portfolio reads. See USE.md.
Quick start (Cursor, self-host)
1. Install
pip install "alloc-context[mcp]"
# From source: pip install -e ".[mcp]"
2. Config and secrets
Copy config/config.example.yaml to
config/config.yaml. Copy .env.example to .env and add
read-only exchange keys when you want portfolio ingest or macro feeds.
See self-hosting.md.
3. MCP config
Add to your global ~/.cursor/mcp.json:
{
"mcpServers": {
"alloc-context": {
"command": "alloc-context",
"args": [
"--config",
"/absolute/path/to/alloc-context/config/config.yaml",
"mcp"
],
"env": {
"ALLOC_CONTEXT_DB": "/absolute/path/to/alloc-context/state/alloccontext.db"
}
}
}
}
Use absolute paths. See cursor-mcp.example.json.
4. Refresh facts (optional)
python -m alloccontext --config config/config.yaml ingest
Run before a session or when you want fresh macro/regime data. No cron required.
5. Ask your agent
Call get_context_bundle for a full snapshot. Pure math tools
(check_allocation_band, get_rebalance_plan) work without portfolio credentials.
Full setup: cursor-mcp.md. Samples: examples.md.
Not financial advice.
MCP tools
| Tool | Purpose |
|---|---|
get_context_bundle | Full ContextBundle — holdings, market, sentiment, macro, delta, regime; optional allocation_analysis |
get_market_context | Sentiment, macro, ETF, breadth, and market fields (no portfolio) |
get_context_at | Saved snapshot from ingest history at a given as_of |
get_context_delta | Notable shifts between two saved snapshots |
get_rebalance_plan | USD rebalance moves from allocation, target, and NAV |
check_allocation_band | Drift vs target and whether allocation is outside the band |
check_allocation_bands | Batch band checks for multiple target scenarios |
get_portfolio_state | Live NAV and holdings (CEX keys or public EVM wallet address) |
get_expectation_review | Score optional local theses against context (pass-through) |
Market context is holdings-scoped: band assets (BTC/ETH) use OHLC bars; alt holdings (e.g. HYPE) use quote snapshots when cached. See context-bundle.md#market-coverage.
See mcp.md for arguments and resources.
Self-host and development
Run ingest and MCP on your machine — the primary supported path.
See self-hosting.md, local-dev.md
(./scripts/dev-up.sh), or docker-self-host.md.
git clone git@github.com:AllocContext/alloc-context.git
cd alloc-context
python3.11 -m venv .venv && source .venv/bin/activate
pip install -e ".[dev,mcp]"
cp .env.example .env
cp config/config.example.yaml config/config.yaml
python -m alloccontext ingest --dry-run
python -m alloccontext rollup --scope daily --stdout
pytest
| Command | Purpose |
|---|---|
python -m alloccontext ingest | Pull configured sources → SQLite |
python -m alloccontext rollup --scope daily --stdout | ContextBundle JSON (facts) |
python -m alloccontext status | Per-source ingest ages, snapshots, MCP /health |
alloc-context mcp | MCP server (stdio or HTTP) |
Optional HTTP MCP + x402 on your host: mcp-http.md.
AllocContext is self-host only — we do not operate mcp.alloc-context.com.
Quickstart: agent-onramp.md · cursor-mcp.md.
Documentation
| Document | Purpose |
|---|---|
| docs/agent-onramp.md | ~2 min quickstart |
| docs/cursor-mcp.md | Cursor stdio MCP |
| docs/self-hosting.md | Local ingest + MCP |
| docs/deterministic-context-mcp-pattern.md | Ingest → rollup → MCP pattern |
| docs/mcp.md | MCP tools |
| docs/examples.md | Sample tool JSON (redacted) |
| docs/context-bundle.md | ContextBundle schema |
| docs/USE.md | License and use policy |
| docs/local-dev.md | Local internal MCP + dev ingest |
| docs/docker-self-host.md | Docker Compose self-host |
| docs/distribution.md | PyPI and MCP Registry |
| docs/publishing.md | Release workflow |
| docs/mcp-http.md | Optional HTTP + x402 (your host) |
| docs/user-config.md | Bridge user.yaml (legacy archive) |
Contributing
GitHub Issues are welcome for bugs, schema feedback, and MCP API suggestions. Unsolicited pull requests are not expected — see CONTRIBUTING.md.
License
MIT License. Self-host via PyPI. See docs/USE.md.
Frequently asked questions
What is AllocContext?
Self-host MCP: portfolio crypto context for agents. PyPI stdio; MIT.
How do I install AllocContext?
It runs on your machine. In Claude Code: claude mcp add alloccontext -- uvx alloc-context==0.3.0 mcp. For other clients, paste the JSON configuration from this page into their MCP settings.
Does AllocContext need an API key?
Its registry entry lists no required key or secret. Some features may still need an account with the service it connects to.
Is AllocContext open source?
Yes. Its code is on GitHub under the MIT license.
Who maintains AllocContext?
It is published by AllocContext. AI.tn lists it but does not maintain or endorse it.
From the official MCP Registry. Listing is not an endorsement: review a server's code and permissions before giving it access to your data.