Philidor MCP
DeFi vault risk analytics for AI agents. Search 700+ vaults across Morpho, Aave, Yearn, Beefy, Spark, and more. Compare risk scores, analyze protocols, run due diligence — all through natural language. No API key required. No installation needed.
Philidor MCP Server
DeFi vault risk analytics for AI agents
Search 700+ DeFi vaults across Morpho, Aave, Yearn, Beefy, and Spark. Compare risk scores, analyze protocols, run due diligence — all through natural language.
No API key required. No installation needed.
Quick Start • Tools • Example Prompts • Risk Framework • Agent Skill
Why Philidor?
Most DeFi data tools give you raw numbers. Philidor gives your AI agent institutional-grade risk intelligence.
| Feature | Philidor | DefiLlama MCP | Generic DeFi APIs |
|---|---|---|---|
| Vault risk scores (0–10) | :white_check_mark: | :x: | :x: |
| Risk vector decomposition | :white_check_mark: | :x: | :x: |
| Vault comparison | :white_check_mark: | :x: | :x: |
| Curator intelligence | :white_check_mark: | :x: | :x: |
| Protocol security history | :white_check_mark: | :x: | Partial |
| Due diligence prompts | :white_check_mark: | :x: | :x: |
| Portfolio risk assessment | :white_check_mark: | :x: | :x: |
| No API key needed | :white_check_mark: | :white_check_mark: | Varies |
| Hosted (zero install) | :white_check_mark: | :x: | :x: |
Quick Start
Remote Server (Recommended)
Connect directly to the hosted server — zero installation, always up to date:
https://mcp.philidor.io/api/mcp
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"philidor": {
"url": "https://mcp.philidor.io/api/mcp"
}
}
}
Claude Code
claude mcp add philidor --transport http https://mcp.philidor.io/api/mcp
Cursor
Add to .cursor/mcp.json:
{
"mcpServers": {
"philidor": {
"url": "https://mcp.philidor.io/api/mcp"
}
}
}
Windsurf
Add to your MCP settings:
{
"mcpServers": {
"philidor": {
"serverUrl": "https://mcp.philidor.io/api/mcp"
}
}
}
Docker (stdio)
docker run -i --rm ghcr.io/philidor-labs/philidor-mcp
Local (stdio)
git clone https://github.com/Philidor-Labs/philidor-mcp.git
cd philidor-mcp
npm install
npm start
Tools
10 tools for vault discovery, risk analysis, and protocol research.
search_vaults
Search and filter DeFi vaults by chain, protocol, asset, risk tier, TVL, and more.
| Parameter | Type | Description |
|---|---|---|
query | string | Search by vault name, symbol, asset, protocol, or curator |
chain | string | Filter by chain (Ethereum, Base, Arbitrum, ...) |
protocol | string | Filter by protocol ID (morpho, aave-v3, yearn-v3, beefy, spark) |
asset | string | Filter by asset symbol (USDC, WETH, ...) |
riskTier | string | Filter by risk tier: Prime, Core, or Edge |
minTvl | number | Minimum TVL in USD |
sortBy | string | Sort field: tvl_usd, apr_net, name |
sortOrder | string | Sort order: asc or desc |
limit | number | Max results (default 10, max 50) |
get_vault
Get detailed information about a specific vault including risk breakdown, recent events, and historical snapshots. Lookup by id or by network + address.
| Parameter | Type | Description |
|---|---|---|
id | string | Vault ID (e.g. morpho-ethereum-0x...) |
network | string | Network slug (ethereum, base, arbitrum) |
address | string | Vault contract address (0x...) |
get_vault_risk_breakdown
Detailed breakdown of a vault's three risk vectors with sub-metrics: asset quality, platform code maturity, and governance controls. Returns dimension-level scores, caps, hard-fail flags, and overrides.
| Parameter | Type | Description |
|---|---|---|
network | string | Network slug |
address | string | Vault contract address |
compare_vaults
Side-by-side comparison of 2–3 vaults on TVL, APR, risk score, risk tier, and audit status.
| Parameter | Type | Description |
|---|---|---|
vaults | array | Array of 2–3 objects with network and address |
find_safest_vaults
Find the top 10 audited, high-confidence vaults sorted by risk score.
| Parameter | Type | Description |
|---|---|---|
asset | string | Filter by asset symbol |
chain | string | Filter by chain name |
minTvl | number | Minimum TVL in USD |
get_protocol_info
Protocol details including TVL, vault count, versions, auditors, bug bounties, and security incidents.
| Parameter | Type | Description |
|---|---|---|
protocolId | string | Protocol ID (morpho, aave-v3, yearn-v3, beefy, spark) |
get_curator_info
Curator details including managed vaults, TVL, chain distribution, and performance metrics.
| Parameter | Type | Description |
|---|---|---|
curatorId | string | Curator ID |
get_market_overview
High-level DeFi vault market statistics: total TVL, vault count, risk distribution, and TVL by protocol. No parameters required.
explain_risk_score
Explain what a specific risk score means, including the tier, calculation method, and thresholds.
| Parameter | Type | Description |
|---|---|---|
score | number | Risk score (0–10) |
list_vaults_with_incidents
List all vaults that had a recent critical incident (last 365 days). Sorted by TVL descending, then by recency. No parameters required.
Resources
| URI | Description |
|---|---|
philidor://methodology | The Vector Risk Framework v4.1 documentation |
philidor://supported-chains | Supported blockchain networks with vault counts |
philidor://supported-protocols | Supported DeFi protocols with TVL data |
Prompts
| Prompt | Description |
|---|---|
vault_due_diligence | Comprehensive due diligence report for a vault |
portfolio_risk_assessment | Portfolio-level risk analysis across positions |
defi_yield_comparison | Yield comparison with risk-adjusted analysis |
Example Prompts
Once connected, try asking your AI assistant:
Discovery
"Find the safest USDC vaults with at least $10M TVL"
"What Morpho vaults are available on Base?"
"Show me the DeFi market overview"
Analysis
"Run due diligence on the Steakhouse USDC vault on Ethereum"
"Compare the top 3 USDC vaults by risk score"
"What's the risk breakdown for this vault: ethereum/0x..."
Portfolio
"Assess my portfolio: 50% in Morpho Steakhouse USDC, 30% in Aave USDC, 20% in Yearn USDC"
"Which protocols have had security incidents?"
"What does a risk score of 8.5 mean?"
Risk Scoring
Philidor uses the Vector Risk Framework v4.1 to decompose vault risk into three measurable vectors:
Final Score = 40% Asset + 40% Platform + 20% Governance
Asset Composition (40%)
Quality of underlying collateral. Blue-chip assets (ETH, USDC) score highest. Factors include oracle reliability, liquidity depth, and peg stability.
Platform Code (40%)
Code maturity measured by:
- Lindy Score — time-based safety (>2 years ≈ 9/10)
- Audit Density — number and quality of audits
- Dependency Risk — multiplicative penalties for risky dependencies
- Incident Penalty — caps score after security incidents
Governance (20%)
Exit window for users:
| Control | Score |
|---|---|
| Immutable contract | 10/10 |
| 7+ day timelock | 9/10 |
| No timelock | 1/10 |
Risk Tiers
| Tier | Score | Meaning |
|---|---|---|
| Prime | 8.0–10.0 | Institutional-grade — mature code, multiple audits, safe governance |
| Core | 5.0–7.9 | Moderate safety — audited but newer or flexible governance |
| Edge | 0.0–4.9 | Higher risk — requires careful due diligence |
Architecture
┌──────────────────┐ ┌─────────────────┐ ┌──────────────┐
│ Claude / Cursor │────▶│ Philidor MCP │────▶│ Philidor API │
│ Windsurf / etc. │◀────│ Server │◀────│ │
└──────────────────┘ └─────────────────┘ └──────┬───────┘
10 tools, 3 resources, │
3 prompts │
┌────▼────┐
│ On-chain │
│ data │
└─────────┘
- Transport: Streamable HTTP (remote) or stdio (local/Docker)
- API: Calls the Philidor Public API — no API key needed
- Stateless: Fresh server instance per request, no session state
- Data: 700+ vaults across Ethereum, Base, Arbitrum, Polygon, Optimism, and Avalanche
Agent Skill
Install the Philidor MCP skill into your coding agent via skills.sh:
npx skills add philidor-labs/philidor-mcp
This gives your agent full knowledge of all tools, resources, prompts, recommended workflows, and best practices for DeFi vault analysis.
Also Available
| Interface | Description | Link |
|---|---|---|
| CLI | Terminal-based vault intelligence — scriptable, pipeable, agent-sandboxed | philidor-cli |
| OpenClaw Skill | Skill definition for the OpenClaw agent platform | npm |
Supported Protocols
Morpho, Aave v3, Yearn v3, Beefy, Spark — with more being added regularly.
See the full list at app.philidor.io.
Development
git clone https://github.com/Philidor-Labs/philidor-mcp.git
cd philidor-mcp
npm install
npm start
The server connects to the public Philidor API by default. To use a custom endpoint:
PHILIDOR_API_URL=http://localhost:3003 npm start
Links
- Philidor Analytics — explore vaults and risk scores
- Philidor CLI — terminal-based vault intelligence
- API Documentation — OpenAPI/Swagger docs
- Risk Methodology — how scores are calculated
- Smithery — MCP server registry
- Twitter — updates and announcements
License
Servidores relacionados
O'RLY MCP
Generates O'RLY? (O'Reilly parody) book covers.
BWA (Burrows-Wheeler Aligner)
An MCP server for the BWA sequence alignment tool.
QMT MCP Server
An MCP server that interfaces with the locally running MiniQMT trading system.
MCP Cat Language Server
A server that translates Chinese text into cute cat language expressions.
CS2 RCON MCP
A server for managing Counter-Strike 2 servers using the RCON protocol.
SwitchBot MCP Server
Control SwitchBot devices interactively using the SwitchBot API.
Robust Long‑Term Memory
A persistent, human‑like memory system for AI companions
Lightning Faucet MCP
Give AI agents a Bitcoin wallet with Lightning Network payments
Context Lens
Semantic search knowledge base for MCP-enabled AI assistants
MCP HUB
The Ultimate Control Plane for MCP Unlock the full power of Model Context Protocol with zero friction. One-Click GPT Integration: Bridge the gap between MCP servers and ChatGPT/LLMs instantly. No more manual config hunting. Pro-Level Orchestration: Manage, monitor, and toggle multiple MCP tools from a single, intuitive dashboard. Secure by Design: Built-in support for complex auth flows and 2FA, making enterprise-grade tool integration seamless. Streamlined Debugging: Test queries and inspect tool responses in real-time without leaving the hub. Stop wrestling with JSON configs. Start building agentic workflows that actually work.