Heavy trading backtests, run in the cloud.
What it does
Runs compute-heavy trading backtests, parameter sweeps, and model training in the cloud and returns performance metrics.
When to use it
Use for multi-year walk-forward tests, large Monte Carlo runs, parameter sweeps, or model training rather than quick local checks.
How to use it
Give it a strategy or model and a ticker symbol in the project; it runs the cloud job and reports trading performance metrics.
What you provide
- a strategy or model
- a ticker symbol
- An existing project
Uses
AM
RU
Good to know
- starts 1 helper agents
- cannot be undone: Anthropic Claude Managed Agent
- works out of sight
Latest review
What you need · 3
Requires Managed Agents beta access.
Requires the MCP tools for managed-agent lifecycle operations, memory storage, and pattern storage.
Access · 4
This skill
Anthropic Claude Managed Agent
Execute
Starts a billed cloud agent (irreversible)
Write
Sends trading jobs to the cloud
memory_store
Write
Stores backtest results and costs