AI task planner that breaks complex requests into steps and delegates to specialized agents, paid per call via x402 on Base.
What it does
Questflow Tars is an AI task planner and coordinator accessible via the x402 payment protocol. It accepts a natural-language message and decomposes complex requests into manageable steps, delegating sub-tasks to specialized agents within the Questflow ecosystem. Payment is handled inline using USDC on the Base network, with a maximum cost of 0.1 USDC per request (the asset contract is 0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913).
The endpoint is a simple HTTP POST that takes a single `input` string field and returns a `response` string. The x402 challenge specifies a maximum timeout of 1200 seconds (20 minutes), suggesting that Tars may orchestrate multi-step workflows that take non-trivial time to complete. The broader Questflow platform focuses on autonomous AI agents ("Clones") that aggregate onchain markets such as Polymarket, Pump.fun, OKX, and Opinion Labs for trading, signal generation, and alpha discovery.
Documentation is sparse: there is no OpenAPI spec, no dedicated docs page, and no agents.txt. The output schema provided in the x402 challenge is the only formal contract available. Users should expect a conversational agent-style interface where the quality and structure of the response depends on the prompt supplied.
Capabilities
Use cases
- —Breaking a complex research or analysis request into sub-tasks handled by specialized agents
- —Coordinating multi-step onchain workflows such as market scanning and trade execution
- —Generating structured action plans from free-form natural-language instructions
- —Delegating heterogeneous tasks (data retrieval, summarization, execution) across agent swarms
Fit
Best for
- —Developers building agent-to-agent pipelines who need a planning layer
- —Users wanting to interact with multiple Questflow agents through a single coordinator
- —Prototyping multi-step AI workflows without managing orchestration logic manually
Not for
- —Production systems requiring deterministic, schema-validated outputs with SLAs
- —Use cases needing sub-second latency (max timeout is 1200 s)
- —Users who need detailed API documentation or guaranteed response formats
Quick start
curl -X POST https://api.questflow.ai/x402/copilot/qrn:copilot:tars \
-H "Content-Type: application/json" \
-H "X-PAYMENT: <x402_payment_header>" \
-d '{"input": "Break down the steps to analyze trending Polymarket events and summarize the top 3 opportunities."}'Example
Request
{
"input": "Break down the steps to analyze trending Polymarket events and summarize the top 3 opportunities."
}Response
{
"response": "Here is your plan:\n1. Scan current Polymarket events sorted by volume and recency.\n2. Filter for events with significant sentiment shifts in the last 24 hours.\n3. Summarize the top 3 opportunities with estimated edge and risk profile."
}Endpoint
Quality
The x402 challenge provides a minimal input/output schema and pricing info, but there is no OpenAPI spec, no dedicated documentation, and no example responses. The response format and capabilities are largely inferred from the description string in the challenge header.
Warnings
- —No OpenAPI spec or formal API documentation available
- —No docs endpoint found (404)
- —Response structure is inferred — only a single 'response' string field is documented
- —Maximum timeout of 1200 seconds suggests potentially long-running requests
- —Example response is fabricated for illustration; actual output format is unknown
Citations
- —x402 challenge specifies maxAmountRequired of 100000 (0.1 USDC) on Base network with asset 0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913https://api.questflow.ai/x402/copilot/qrn:copilot:tars
- —Description states Tars is an AI task planner and coordinator that breaks down complex requests into manageable steps and delegates to specialized agentshttps://api.questflow.ai/x402/copilot/qrn:copilot:tars
- —maxTimeoutSeconds is 1200https://api.questflow.ai/x402/copilot/qrn:copilot:tars
- —Questflow platform aggregates Polymarket, Pump.fun, OKX, and Opinion Labshttps://api.questflow.ai