Calibrated probabilistic foresight for AI agents, powered by live prediction-market signal.
Scored across 2 tools
The two tools have clearly distinct purposes: one for synthesizing a probability forecast from evidence, the other for discovering relevant markets. Their descriptions explicitly state when to use each, eliminating ambiguity.
Both tools follow the same 'polybridge_' prefix followed by a descriptive verb-noun pattern ('search' and 'forecast'), creating a predictable and consistent naming convention.
With only two tools, the set is minimal but appropriately scoped for discovery and forecasting. It could benefit from one or two additional tools (e.g., market detail lookup) but remains effective for its stated purpose.
The tools cover the core workflow of searching markets and generating forecasts, but lack direct access to individual market details or historical data, which would enhance completeness for a prediction market assistant.
Generate a read-only probability forecast for a clearly stated future event by searching relevant prediction markets and synthesizing evidence. Use when the user asks for a probability, outlook, or forecast; use polybridge_search when they only need market discovery. Does not place trades, provide financial advice, or access private/internal data.
| Name | Required | Description | Default |
|---|---|---|---|
| question | Yes | Clearly stated future-event forecasting question, 500 characters or fewer. | |
| include_graph | No | Whether to include the causal graph in the structured response. |
| Name | Required | Description |
|---|---|---|
| error | No | |
| metadata | No | |
| question | Yes | |
| reasoning | Yes | |
| confidence | Yes | |
| latency_ms | Yes | |
| request_id | Yes | |
| engine_type | Yes | |
| probability | No | |
| causal_graph | No | |
| distribution | No | |
| markets_used | Yes | |
| confidence_interval | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint and openWorldHint. The description adds further behaviors: does not place trades, provide financial advice, or access private/internal data, with no contradiction.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, front-loaded with the core purpose and usage guidance, no redundant information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a tool with output schema and annotations, the description covers purpose, usage, and limitations completely. No missing context.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so baseline 3. The description does not add extra parameter-level detail beyond what the schema already provides.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool generates a read-only probability forecast by searching markets and synthesizing evidence, and explicitly distinguishes it from the sibling polybridge_search.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly says when to use the tool (user asks for probability/outlook/forecast) and when not (use polybridge_search for market discovery only), and clarifies limitations like no trades or financial advice.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search public prediction markets for markets relevant to a natural-language topic or question. Use when you need candidate markets, market URLs, outcomes, statuses, and relevance scores. Do not use for a final probability forecast, market-history lookup, trading, or private/internal data. Scores are relevance scores, not probabilities.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | Natural-language topic or question to match to public markets. | |
| filters | No | Optional market filters. Currently only filters.status is supported, with active, closed, or resolved. | |
| dimensions | No | Search dimensions to run. Uses all supported dimensions by default. | |
| top_k_per_dimension | No | Maximum number of candidate markets to return per search dimension. |
| Name | Required | Description |
|---|---|---|
| query | Yes | |
| results | Yes | |
| warnings | No | |
| request_id | Yes | |
| total_markets | Yes | |
| dimensions_returned | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already indicate readOnlyHint and openWorldHint. Description adds context about public vs private data and clarifies score interpretation, but could mention rate limits or authentication needs. No contradiction with annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three sentences, front-loaded with core purpose, no wasted words. Every sentence adds value.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With an output schema present, the description does not need to detail return values. It covers the purpose, use cases, exclusions, and score interpretation, making it complete for a search tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% and parameters are well-described in the schema. The description does not add significant extra meaning beyond what the schema already provides, so baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'Search' and the resource 'public prediction markets', specifies the returned fields (candidate markets, URLs, outcomes, statuses, relevance scores), and distinguishes from the sibling tool polybridge_forecast.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Explicitly states when to use (need candidate markets) and when not to use (final probability forecast, market-history lookup, trading, private data). Also clarifies that scores are relevance scores, not probabilities.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
polybridge_forecastpolybridge_searchCalibrated world model for AI agents. 40 tools: world state, markets, trading. Kalshi + Polymarket.
Machine-payable prediction-market briefs and probability-change monitoring for AI agents.
Financial data for AI agents: crypto data, Polymarket odds, weather/oil calibration, trust scoring.
Live trading-pipeline intelligence for AI agents: signal scoring, calibration, recorded outcomes.
Add one secure layer between your agents and this server.
