Which integrations are available for this server?

Integrates Stripe for processing payments and subscriptions for credit packs and premium tiers.

How do I use Sablier MCP Server?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Sablier MCP Server How exposed is my portfolio to recession risk?" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

Sablier MCP Server

Official

by sablier-ai

Overview Schema Related Servers Score Discussions

Python

Hybrid

Sablier MCP Server

License: MIT Python 3.11+ MCP 1.26+

An MCP (Model Context Protocol) server that lets AI assistants analyze portfolios, stress-test scenarios, generate synthetic market paths, and scan SEC filings — in under 2 minutes.

What This Does

Connect Sablier to Claude, ChatGPT, or any MCP-compatible AI assistant. The agent gets ~70 tools to:

Scan SEC filings & earnings calls — AI reads every company's 10-K, 10-Q filings and earnings call transcripts, then scores how exposed each holding is to any theme you ask about (0-100 scale with evidence). "How exposed is my portfolio to China supply chain risk?" — scored, evidenced, and ranked in seconds.

Compute factor exposures — Measures how each stock responds to market drivers (interest rates, VIX, dollar index, oil, credit spreads, etc.). Factor betas are estimated on a rolling window of recent data so they reflect current market conditions. Supports both linear and nonlinear (GAM) factor models.

Stress-test with scenarios — "What if VIX hits 40?" or "What if the Fed raises rates to 6%?" Run Monte Carlo simulations to get per-asset expected returns, Value-at-Risk, Expected Shortfall, and full return distributions.

Generate synthetic market paths — Train a generative flow model on the joint distribution of assets and factors. Generate hundreds of realistic future trajectories, optionally constrained ("paths where gold stays above $3000"). Forward-test strategies across many scenarios.

Real-time market intelligence — Get a Bloomberg-terminal-grade briefing with 50+ indicators, z-scores, regime signals, and cross-asset analysis in one call.

Manage portfolios — Create and track portfolios with live prices, performance analytics (Sharpe ratio, max drawdown, volatility), optimization, and efficient frontier computation.

Speed

Under 2 minutes, end-to-end. Portfolio creation -> model training -> factor betas -> stress scenarios -> SEC filing analysis. All in a single conversation.
The same workflow — gathering filings, building factor models, running simulations, writing risk memos — takes a team of analysts and quants days to weeks. Sablier compresses it into one chat.

Related MCP server: SEC EDGAR MCP

Quick Start

Option A: Claude Desktop / Claude.ai (recommended — zero install)

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "sablier": {
      "type": "url",
      "url": "https://sablier-mcp-215397666394.us-central1.run.app/mcp/"
    }
  }
}

That's it. On first use, Claude opens a browser window — sign up or log in with your Sablier account. No API keys needed.

Option B: ChatGPT (Developer Mode)

Go to Settings -> Developer Mode and enable it
Go to Connectors -> Add connector
Enter:
- Name: Sablier
- Server URL: https://sablier-mcp-215397666394.us-central1.run.app/mcp/
In a new chat, activate via "+" -> More -> Developer Mode

Option C: Claude Code (local, stdio)

# Clone and install
git clone https://github.com/sablier-it/sablier-mcp.git
cd sablier-mcp
uv sync

# Register with Claude Code
claude mcp add sablier -- uv --directory /path/to/sablier-mcp run sablier-mcp

# Set your API key (get one from sablier-ai.com)
export SABLIER_API_KEY=sk_live_your_key_here

Tools (~70)

Market Intelligence

Tool	Description
`market_radar`	Bloomberg-grade briefing: 50+ indicators, z-scores, regime signals, cross-asset analysis
`search_features`	Search for tickers (stocks, ETFs, futures) and market indicators (VIX, DXY, rates)

Portfolio Management

Tool	Description
`create_portfolio`	Create a portfolio from tickers and weights (must sum to 1.0)
`list_portfolios`	List your portfolios with names, assets, and status
`get_portfolio`	Get full details for a specific portfolio
`update_portfolio`	Update name, description, weights, or capital
`delete_portfolio`	Permanently delete a portfolio
`get_portfolio_value`	Live portfolio value: total, P&L, per-position breakdown
`get_portfolio_analytics`	Sharpe ratio, volatility, max drawdown, beta (1W-5Y timeframes)
`get_asset_profiles`	Sector, industry, country, and exchange for each holding
`optimize_portfolio`	Find optimal weights: max Sharpe, min variance, or max return
`get_efficient_frontier`	Compute the efficient frontier curve for portfolio assets

Qualitative Analysis (SEC Filings & Earnings Calls)

Tool	Description
`analyze_qualitative`	Score company exposure to any theme (0-100) using 10-K, 10-Q, and earnings transcripts
`list_themes`	Browse the built-in theme library (AI risk, rate sensitivity, China exposure, etc.)
`list_grain_analyses`	List past qualitative analyses
`get_grain_analysis`	Load a saved analysis with full scores and evidence passages
`delete_grain_analysis`	Delete a saved qualitative analysis

Quantitative Analysis (Factor Models)

Tool	Description
`analyze_quantitative`	One-shot: builds factor models, trains, computes factor exposures (linear + nonlinear)
`list_model_groups`	List existing analyses with training and simulation status
`list_feature_set_templates`	Browse pre-built market driver sets (rates, volatility, commodities, credit, etc.)
`create_feature_set`	Create a custom set of market drivers for analysis
`list_feature_sets`	List all accessible feature sets
`get_feature_set`	Get details of a specific feature set
`delete_feature_set`	Delete a custom feature set
`compute_betas`	Compute per-asset factor betas from a trained model group
`compute_returns`	Sample forward asset returns under user-specified factor levels (Monte Carlo)
`get_residual_correlation`	Cross-asset residual correlation matrix (post-factor)
`list_simulations`	List all simulations for a model group
`optimize_portfolio`	Find optimal weights from a trained MOMENT batch (analytical objectives)
`get_optimization_history`	Browse past optimization runs for a portfolio
`delete_model_group`	Delete a model group and all associated data

Stress Testing & Scenarios

Tool	Description
`compute_returns`	Monte Carlo what-if: per-asset VaR, ES, expected return under custom factor levels
`create_scenario`	Save a named what-if scenario (fixed value, percentile, or shock)
`list_scenarios`	List saved scenarios
`get_scenario`	Get scenario details
`update_scenario`	Update a scenario's factors or description
`delete_scenario`	Delete a saved scenario

Generative Simulation (Flow)

Tool	Description
`train_flow_model`	Train a conditional flow-matching model on a model group
`generate_flow_paths`	Generate baseline (unconstrained) synthetic price paths
`simulate_flow_scenario`	Generate constrained paths (e.g., "gold above $3000 and VIX below 20")
`check_flow_job`	Poll status of a queued/running flow job
`get_flow_results`	Pull paths + summary stats from a completed generation job
`download_flow_paths`	Download raw paths as JSON or CSV
`list_flow_baselines`	List baseline path generations for a model group
`list_flow_scenarios`	List scenario-conditioned generations for a model group
`delete_flow_job`	Delete a flow job and its outputs
`test_flow_risk`	Compute portfolio risk metrics across generated paths
`flow_validate`	Validate flow model quality against historical data (calibration, dependence, tails)

Feature Catalog

Tool	Description
`search_features`	Search the catalog for available tickers / macro series
`add_feature`	Add a ticker to the catalog (Yahoo Finance or FRED)
`refresh_feature_data`	Fetch/update historical data for tickers

Trading Rules

Tool	Description
`create_rule`	Create a trigger-action rule on a portfolio (VIX spike, drawdown, regime, etc.)
`list_rules`	List rules attached to a portfolio
`toggle_rule`	Activate or deactivate a rule
`delete_rule`	Permanently delete a rule
`evaluate_rules`	Check which rules would fire on the latest market data
`backtest_rules`	Historical backtest of rules against real prices
`forward_test_rules`	Forward-test rules against generated FLOW paths

Derivatives

Tool	Description
`analyze_derivatives`	Greeks aggregation + scenario P&L for portfolio options positions
`price_option_tool`	Price a single equity / futures option (Black-76) with Greeks

Market Data

Tool	Description
`market_radar`	60+ regime indicators (VIX, yield curve, credit, sector rotation, RORO)
`screen_universe`	Cross-sectional screening across the equity catalog

Account & Billing

Tool	Description
`whoami`	Current user, tier, and feature flags
`get_credits`	Current credit balance and tier monthly allocation
`list_credit_packs`	Browse one-off credit packs available for purchase
`buy_credit_pack`	Purchase a credit pack via Stripe Checkout
`get_billing_info`	View current subscription tier and limits
`get_billing_usage`	View usage across metered buckets
`toggle_overage`	Enable/disable per-call overage spending
`subscribe`	Subscribe or upgrade via Stripe Checkout
`manage_subscription`	Open Stripe Customer Portal

Example Conversations

1. Full risk analysis in one conversation

You:   Create a portfolio with AAPL 40%, MSFT 30%, NVDA 30%.
       Then stress-test it for a recession — VIX at 35, 10Y at 5.5%, SPY at 380.

Agent: 1. create_portfolio("Tech Portfolio", ["AAPL", "MSFT", "NVDA"], [0.4, 0.3, 0.3])
       2. list_feature_set_templates()  →  picks "Macro + Volatility" set
       3. analyze_quantitative(portfolio_id, conditioning_set_id)
          →  trains models, computes factor betas per asset
       4. compute_returns(sim_batch_id, {"VIX": 35, "US 10Y": 5.5, "SPY": 380})
          →  per-asset expected returns, VaR (95%), Expected Shortfall

       Result: Portfolio expected return = -8.2%, VaR(95%) = -14.5%
       NVDA most exposed (-12.1%), MSFT most defensive (-4.8%)

2. SEC filing analysis for thematic risk

You:   How exposed are the Magnificent 7 to AI regulation risk?

Agent: 1. analyze_qualitative(
            tickers=["AAPL", "MSFT", "GOOGL", "AMZN", "NVDA", "META", "TSLA"],
            themes=["AI regulation risk"]
       )
       2. Returns per-ticker scores (0-100) with evidence from 10-K filings:
          - META: 82/100 (HIGH) — "government regulation of AI... could limit our ability
            to deploy AI features across our family of apps"
          - GOOGL: 78/100 — "new AI regulations could require us to restrict or modify
            products and services"
          - NVDA: 71/100 — "export controls... restrictions on AI chip sales to China"
          - TSLA: 31/100 (LOW) — limited AI regulation mentions in filings

3. Generative scenario analysis

You:   I want to see how a gold + bonds portfolio performs over the next quarter
       in scenarios where inflation stays high.

Agent: 1. create_portfolio("Inflation Hedge", ["GLD", "TLT", "IAU"], [0.4, 0.4, 0.2])
       2. train_flow_model(model_group_id, horizon=60)
          →  dispatches GPU training job (~5-15 min async)
       3. check_flow_job(job_id) until completed; flow_validate(model_group_id)
          →  confirms model quality before generating paths
       4. generate_flow_paths(model_group_id, n_paths=500)
          →  generates 500 joint baseline price paths
       5. simulate_flow_scenario(model_group_id, constraints=[
            {"feature_name": "CPI", "type": "level", "lower": 3.5, "t_start": 0, "t_end": 60}
          ])
          →  generates paths conditioned on CPI > 3.5%
       6. test_flow_risk(portfolio_id, job_id)
          →  distribution of Sharpe, max drawdown, total return across constrained paths

       Result: Median return = +4.2%, 5th percentile = -6.8%
       Gold outperforms bonds in 72% of high-inflation paths

4. Market briefing and portfolio checkup

You:   What's happening in markets today and how is my portfolio positioned?

Agent: 1. market_radar()
          →  50+ indicators: equities, rates, credit, FX, commodities, volatility
          →  regime signals: Risk-Off score = 0.7, yield curve inverted, VIX elevated
       2. list_portfolios()  →  finds your "Tech Portfolio"
       3. get_portfolio_value(portfolio_id)
          →  current value, daily P&L, per-position breakdown
       4. get_portfolio_analytics(portfolio_id, timeframe="1M")
          →  1-month Sharpe, volatility, max drawdown

       Result: Markets are risk-off (VIX +15% this week, credit spreads widening).
       Your tech portfolio is down -2.3% today, concentrated in high-beta names.
       Consider: stress-test with compute_returns to quantify downside risk.

Authentication

Remote mode (Claude Desktop, Claude.ai, ChatGPT): OAuth 2.0 browser-based login — no API keys to manage. Sablier authenticates via Google OAuth, then issues an API key for the MCP session.
Local mode (Claude Code, stdio): Set SABLIER_API_KEY environment variable.

Pricing

Tier	Price	Included
Free	$0	10 market radar, 5 factor models, 2 GRAIN analyses/mo
Pro	$79/mo	100 market radar, 50 factor models, 20 GRAIN, 10 Flow sims/mo
Enterprise	$399/mo/seat	Unlimited everything, priority support

Portfolio management, read operations, and scenario management are always free. Overages billed per-call beyond included limits.

Architecture

sablier-mcp/
├── src/sablier_mcp/
│   ├── server.py      # 53 MCP tool definitions (FastMCP)
│   ├── client.py      # Async HTTP client for Sablier API
│   ├── auth.py        # OAuth 2.0 provider (remote mode)
│   └── widgets.py     # Rich HTML cards for Claude Desktop
├── pyproject.toml
├── Dockerfile
└── README.md

Remote mode (Claude Desktop, Claude.ai, ChatGPT): OAuth 2.0 browser login — no API keys to manage
Local mode (Claude Code, stdio): API key from environment variable
Widgets: Tools return rich HTML cards (beta heatmaps, score cards, portfolio overviews) alongside text for visual output in Claude Desktop

Development

# Run the server locally (stdio transport)
uv run sablier-mcp

# Test with MCP inspector
npx @modelcontextprotocol/inspector uv --directory . run sablier-mcp

# Run as remote server (streamable-http with OAuth)
MCP_TRANSPORT=streamable-http uv run sablier-mcp

Privacy Policy

Sablier processes portfolio data and market queries to provide analytics. Full details:

Data collected: Portfolio holdings, factor model parameters, and query metadata for analytics computation
Usage & storage: Data is processed on Sablier's servers (GCP, US) and stored for your account's analytics history. Portfolios and models persist until you delete them.
Third-party sharing: Sablier does not sell or share your data. Market data is sourced from public feeds. SEC filings are public records.
Retention: Account data is retained while your account is active. Deleted portfolios and models are purged within 30 days.
Contact: team@sablier.it

Full privacy policy: sablier-ai.com/privacy | Terms of service: sablier-ai.com/terms

Support

Email: team@sablier.it
Issues: github.com/sablier-it/sablier-mcp/issues

Links

Sablier Platform: sablier-ai.com
MCP Protocol: modelcontextprotocol.io

Install Server

license - permissive license

quality

maintenance

How are these scores calculated?

Maintenance

–Maintainers

–Response time

–Release cycle

–Releases (12mo)

Commit activity

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Tools

View all tools

Related MCP Servers

AdvisorFinder MCP Server
Finance Search Government Data
drew-keever
A
license
A
quality
B
maintenance
An MCP server that gives AI assistants access to SEC-registered financial advisor data, enabling searches by name, firm, or state, full profile lookups, disclosure history checks, and risk assessments.
Updated 4 days ago (2026-08-12 20:43 UTC)
6
1
MIT
SEC EDGAR MCP
Finance Government Data
stefanoamorelli
A
license
C
quality
C
maintenance
MCP server for accessing SEC EDGAR filings. Connects AI assistants to company filings, financial statements, and insider trading data with exact numeric precision.
Updated 4 days ago (2026-08-12 20:16 UTC)
21
348
AGPL 3.0
CyberQuant MCP Server
Finance Data Platforms
slb-cn
A
license
A
quality
B
maintenance
MCP server for CyberQuant data sharing platform that enables AI assistants to directly query and analyze financial data using CSV output for token efficiency.
Updated a month ago (2026-07-17 08:23 UTC)
7
28
MIT
cvi-mcpofficial
Finance Research & Data
Computational-Value-Investing
A
license
-
quality
C
maintenance
MCP server that makes the CVI Platform a native tool for AI assistants, enabling stock screening, company analysis, and point-in-time research via natural language.
Updated a month ago (2026-07-16 12:40 UTC)
MIT

View all related MCP servers

Related MCP Connectors

Drillr — The financial MCP for AI agents
The financial MCP for AI agents - 90+ financial tables, SEC filings, signals, alt-data.
mcp
MCP server connecting AI agents to non-custodial staking data across 130+ networks.
ko-mcp
Real SEC, 13F, insider, congress & macro data your AI agent can cite. Hosted MCP, 24 tools.

View all MCP Connectors

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/sablier-ai/sablier-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server

Sablier MCP Server

What This Does

Speed

Quick Start

Option A: Claude Desktop / Claude.ai (recommended — zero install)

Option B: ChatGPT (Developer Mode)

Option C: Claude Code (local, stdio)

Tools (~70)

Market Intelligence

Portfolio Management

Qualitative Analysis (SEC Filings & Earnings Calls)

Quantitative Analysis (Factor Models)

Stress Testing & Scenarios

Generative Simulation (Flow)

Feature Catalog

Trading Rules

Derivatives

Market Data

Account & Billing

Example Conversations

1. Full risk analysis in one conversation

2. SEC filing analysis for thematic risk

3. Generative scenario analysis

4. Market briefing and portfolio checkup

Authentication

Pricing

Architecture

Development

Privacy Policy

Support

Links

Maintenance

Resources

Looking for Admin?

Tools

Related MCP Servers

AdvisorFinder MCP Server

SEC EDGAR MCP

CyberQuant MCP Server

cvi-mcpofficial

Related MCP Connectors

Latest Blog Posts

MCP directory API