--- url: "https://predictiontoolkit.com/tools/simmer-markets/" publisher: "PredictionToolkit" dateModified: "2026-11-03" --- [← AI agents]() ![Simmer logo]() Platform # Simmer Simmer is a hosted platform built around reusable trading skills for AI agents. Its catalog covers crypto, weather, copying and other strategies; the builder can turn an idea into a skill containing configuration and a trading script. [**3.5**/ 5Editorial rating]() [Editorial assessment]() [Explore Simmer skills]() simmer.markets **Available on:** Web · CLI · MCP **Pricing:** [See plans and costs ↓]() [Screenshots]() [Explore alternatives]() [AI agents]() [Strategy builders]() +4 more [Trading bots]() [Paper trading]() [MCP servers]() [Copy trading]() ## Screenshots and interface Product preview1 / 2 The provider’s published skill catalog. ## What is Simmer? Skills define market selection and strategy behavior, while account-level trade, daily and drawdown caps sit outside the strategy’s own decisions. The workflow starts with play money and keeps live-wallet linking separate. The provider says the signing key remains with the agent, and trade records expose the reasoning attached to its decisions. The provider lists CLI agents, ClawHub and MCP access, with Polymarket, Kalshi, Hyperliquid and simulated markets in the venue coverage. Pro labels apply to Autoresearch experiments using a holdout split and filtered real-time whale or copying signals delivered to the agent. This profile covers the hosted skill platform; the developer SDK is a separate offering. ### Who it is for - For developers running research agents and users adapting a documented skill to their own configuration. - For researchers comparing an agent’s reasoning, simulated decisions and settled results. ## Markets and platform rules - **Catalog and skill builder** Install a strategy skill or describe an idea to generate configuration and a trading script, then inspect how it selects markets and executes. - **Paper start and external caps** Use play money before live-wallet linking, apply account-level trade, daily and drawdown caps and inspect decision records. Local signing is described as keeping the key with the agent. - **Pro experiments and signals** Autoresearch uses a holdout split to assess experiments; Pro signals deliver filtered whale and copying activity to the agent. Subscription and execution costs require account-level confirmation. ## Market participation workflows ### Read before running a skill Inspect market universe, trigger, sizing, exits and dependencies, then keep the default dry-run or play-money mode while reviewing decisions. ### Compare a generated variant Describe the variation you want, then read the generated configuration and script to see what changed. Run the variant with simulated funds and compare its decisions with the original settings before drawing conclusions about the change. ### Separate live activation Treat live activation as a separate decision after simulation. Check the linked wallet and the account’s trade, daily and drawdown caps, then inspect what grants permission to submit real orders before enabling that permission. ## Pricing and total cost The provider describes catalog skills as free to install. Autoresearch and real-time signals are marked Pro; the reviewed materials do not establish a complete current subscription or execution charge, so check the account terms. ## Access and participation 1. Choose an agent and CLI, ClawHub or MCP integration, then read the skill and its dependencies. 2. Keep dry-run or play money, set limits and inspect reasoning and simulated positions. 3. Link a wallet only for a separately chosen live workflow after reviewing activation and costs. ## Useful links ### What to check - Simulated and live venues are separate modes. Changes to the model or configuration can change market selection and decisions. - Installing a skill does not demonstrate that its strategy works in your configuration. A closer look ## Ratings & reviews Editorial rating **3.5**/ 5 0 user reviews PredictionToolkit editorial ### Our take on Simmer The skill catalog, paper start and account-level caps provide a useful experimental path. A skill installation does not validate a strategy, and Pro or execution costs still need checking in the account. **Assessment criteria:** Task usefulness, documented scope, setup clarity, and access and cost transparency. **Assessment date:** 2026-10-05 **Documentation reviewed:** 2026-10-01 Documentation-based assessment. [How we rate]() ### User reviews 0 The main score switches to user ratings after 5 published reviews. **Be the first to share your experience** A specific example helps the next person choose the right tool. ## Related tools [Browse category]() ![Ask Gina research and wallet assistant with embedded demonstration]() ![Ask Gina logo]()[Ask Gina]() Platform**3.5**Editorial Research markets conversationally, use supported order workflows and schedule wallet tasks with explicit access choices. - Access Web · MCP Also in AI Agents and Assistants. ![Prophed strategy-builder preview with backtest settings]() ![Prophed logo]()[Prophed]() Platform**3.5**Editorial Develop prediction-market rules with explicit costs, risk limits, historical replay and a paper stage. - Trading mode Paper · Live automation - Access Web Compare an inspectable strategy builder. ![Simmer SDK introduction and integration paths]() ![Simmer SDK logo]()[Simmer SDK]() SDK**3.5**Editorial Develop agent strategies with research, paper trading, replay and supported live integrations. - Language Python - Access CLI Explore the underlying developer SDK. ## Frequently asked questions about Simmer ### Is this the developer SDK? This profile covers the hosted skill platform; the underlying SDK has a separate profile. ### Where does signing happen? The provider describes local signing with the key remaining with the agent. In this profile