Skip to main content
Glama
Sign In
vynn-mcp

vynn-mcp

by beee003
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

vynn-mcp

mcp-name: io.github.beee003/vynn-mcp

MCP server for Vynn — self-improving AI workflows & backtesting.

Tools

Workflows

  • list_workflows — List your workflows

  • get_workflow — Get workflow details with steps

  • create_workflow — Create a new workflow

  • run_workflow — Execute a workflow

  • get_runs / get_run_summary — View run history

Self-Improving

  • optimize_prompt — AI-powered prompt optimization for a step

  • apply_prompt_optimization — Apply an optimized prompt

  • get_model_recommendation — Get model swap recommendations

  • set_schedule / get_schedule / delete_schedule — Cron scheduling

  • create_trigger / list_triggers — Inbound webhook triggers

Backtesting

  • backtest — Run a backtest with natural language or structured strategies

  • batch_backtest — Parameter sweep across strategy variations

  • optimize_portfolio — Mean-variance portfolio optimization

Utilities

  • list_templates / clone_template — Workflow templates

  • list_available_tools — Tools attachable to workflow steps

  • get_analytics — Workflow performance analytics

Related MCP server: knitbrain

Setup

# Install
pip install vynn-mcp

# Or from source
pip install -e /path/to/vynn-mcp

Set your API key:

export VYNN_API_KEY="vynn_free_..."

Usage with Claude Code

Add to your Claude Code MCP config (~/.claude.json):

{
  "mcpServers": {
    "vynn": {
      "command": "vynn-mcp",
      "env": {
        "VYNN_API_KEY": "vynn_free_..."
      }
    }
  }
}

Usage with Cursor

Add to .cursor/mcp.json:

{
  "mcpServers": {
    "vynn": {
      "command": "vynn-mcp",
      "env": {
        "VYNN_API_KEY": "vynn_free_..."
      }
    }
  }
}

Get an API Key

Sign up at the-vynn.com or:

curl -X POST https://the-vynn.com/v1/signup \
  -H "Content-Type: application/json" \
  -d '{"email": "you@example.com"}'

Available Tools

21 tools
apply_prompt_optimizationC

Apply an optimized prompt to a workflow step.

Args: workflow_id: UUID of the workflow step_id: UUID of the step prompt: The new prompt text to apply

Inputs: workflow_id (required), step_id (required), prompt (required). Returns: result

backtestA

Run a backtest on a trading strategy using natural language or structured rules.

Args: strategy: Strategy description (e.g. "Buy when RSI(14) < 30, sell when RSI(14) > 70") or JSON structured format {"entries":[...],"exits":[...]} universe: List of tickers (e.g. ["AAPL", "MSFT"]) or preset ("SP500", "MAGNIFICENT7"). Defaults to MAGNIFICENT7. period_start: Start date YYYY-MM-DD (default: 3 years ago) period_end: End date YYYY-MM-DD (default: today) capital: Starting capital (default: 100000) benchmark: Benchmark ticker (default: SPY) walk_forward: Enable walk-forward validation to detect overfitting position_sizing: "equal", "risk_parity", "vol_target", or "kelly" allow_short: Allow short selling

Inputs: strategy (required), universe, period_start, period_end, +5 more. Returns: result

batch_backtestA

Run a parameter sweep across strategy variations and rank them.

Args: base_strategy: Strategy template with {placeholders} (e.g. "Buy when RSI({period}) < {threshold}") parameter_grid: Dict of param name to list of values (e.g. {"period": [14, 21], "threshold": [25, 30, 35]}) universe: List of tickers or preset period_start: Start date YYYY-MM-DD period_end: End date YYYY-MM-DD capital: Starting capital rank_by: Metric to rank by — "sharpe_ratio", "total_return_pct", or "max_drawdown_pct"

Inputs: base_strategy (required), parameter_grid (required), universe, period_start, +3 more. Returns: result

clone_templateA

Clone a workflow template into your account.

Args: template_id: UUID of the template name: Name for your new workflow

Inputs: template_id (required), name (required). Returns: result

create_triggerA

Create an inbound webhook trigger for a workflow.

Returns a URL and secret that external services can POST to in order to fire the workflow.

Args: workflow_id: UUID of the workflow name: Human-readable name for the trigger

Inputs: workflow_id (required), name. Returns: result

create_workflowA

Create a new Vynn workflow.

Args: name: Workflow name (e.g. "Daily Market Brief") description: What this workflow does steps: Optional list of step dicts, each with 'name' and 'description' (the system prompt). Example: [{"name": "Research", "description": "Find top news for {topic}"}] email_recipients: Optional list of email addresses to send workflow output to after each execution. Example: ["maya@gmail.com", "bob@example.com"] email_subject: Optional custom email subject. Defaults to "Vynn: {workflow_name}".

Inputs: name (required), description (required), steps, email_recipients, +1 more. Returns: result

delete_scheduleB

Remove the schedule from a workflow.

Args: workflow_id: UUID of the workflow

Inputs: workflow_id (required). Returns: result

get_analyticsB

Get performance analytics for a workflow — success rates, latency, cost, model usage.

Args: workflow_id: UUID of the workflow

Inputs: workflow_id (required). Returns: result

get_model_recommendationA

Get AI-powered model recommendation for a workflow step based on performance data.

Args: workflow_id: UUID of the workflow step_id: UUID of the step

Inputs: workflow_id (required), step_id (required). Returns: result

get_runsB

Get recent run history for a workflow.

Args: workflow_id: UUID of the workflow limit: Number of runs to return (default 10)

Inputs: workflow_id (required), limit. Returns: result

get_run_summaryB

Get detailed summary of a specific workflow run.

Args: run_id: UUID of the run

Inputs: run_id (required). Returns: result

get_scheduleA

Get the current schedule for a workflow.

Args: workflow_id: UUID of the workflow

Inputs: workflow_id (required). Returns: result

get_workflowA

Get full details of a workflow including all steps and their configuration.

Args: workflow_id: UUID of the workflow

Inputs: workflow_id (required). Returns: result

list_available_toolsA

List all tools that can be attached to workflow steps (web search, code exec, etc).

Inputs: no named parameters. Returns: result

list_templatesA

List available workflow templates that you can clone.

Inputs: no named parameters. Returns: result

list_triggersB

List all webhook triggers for a workflow.

Args: workflow_id: UUID of the workflow

Inputs: workflow_id (required). Returns: result

list_workflowsA

List all your Vynn workflows with their steps and status.

Inputs: no named parameters. Returns: result

optimize_portfolioA

Optimize portfolio weights using mean-variance optimization.

Args: universe: List of tickers (e.g. ["AAPL", "MSFT", "GOOGL", "TSLA"]) method: "max_sharpe", "min_variance", "risk_budget", or "black_litterman" period_start: Start date YYYY-MM-DD period_end: End date YYYY-MM-DD long_only: Only allow long positions (default: true)

Inputs: universe (required), method, period_start, period_end, +1 more. Returns: result

optimize_promptA

Analyze recent failures and generate optimized prompt variants for a workflow step.

Uses LLM-powered analysis of inference logs to suggest better prompts. Returns up to 5 variants ranked by estimated improvement.

Args: workflow_id: UUID of the workflow step_id: UUID of the step to optimize

Inputs: workflow_id (required), step_id (required). Returns: result

run_workflowB

Execute a workflow with the given input and return results.

Args: workflow_id: UUID of the workflow to run input_text: The input text/prompt to feed into the workflow

Inputs: workflow_id (required), input_text (required). Returns: result

set_scheduleA

Schedule a workflow to run automatically on a cron schedule.

Args: workflow_id: UUID of the workflow cron_expression: Cron expression (e.g. "0 8 * * 1-5" for weekdays at 8am) default_input: The input text to use for each scheduled run timezone: Timezone for the schedule (default: UTC)

Inputs: workflow_id (required), cron_expression (required), default_input (required), timezone. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    B3.3/5.0

    Scored across 21 tools

    Disambiguation4/5

    Most tools target distinct actions or resources (e.g., create_workflow vs list_workflows, backtest vs batch_backtest). Minor overlap between optimize_prompt (generates variants) and apply_prompt_optimization (applies a prompt) but descriptions clarify the difference.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun pattern in snake_case (e.g., create_workflow, set_schedule, list_triggers). No mixed conventions or irregular names.

    Tool Count3/5

    21 tools is above the typical well-scoped range (3-15). While each tool has a clear purpose, the number feels slightly heavy for a single server, especially covering both workflow management and trading optimization.

    Completeness3/5

    Workflow management lacks delete_workflow and update_workflow, leaving gaps in lifecycle coverage. Trading tools cover backtesting and optimization but are limited compared to a full trading platform. The set is functional but not fully comprehensive.

    Maintenance

    ActivityInactive
    ResponsivenessLimited data

    Resources

    Related MCP Connectors

    Related MCP Servers

    • Tyra Advanced Memory MCP

      bigrock776
      -
      license
      Not graded
      quality
      Not graded
      maintenance
      A sophisticated MCP server providing advanced memory capabilities with RAG, hallucination detection, and enterprise-grade AI infrastructure for intelligent agent ecosystems.
      -

    • knitbrain

      PDgit12
      A
      license
      A
      quality
      C
      maintenance
      Local-first MCP server that gives any AI coding agent per-project memory, workflow intelligence, and always-on, lossless token & context optimization.
      37
      65 npm
      6
      MIT