AI task planner that breaks complex requests into steps and delegates to specialized agents, paid per call via x402 on Base Sepolia.
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 returns a structured response after breaking down the request into manageable steps and delegating sub-tasks to specialized agents. The endpoint is live on Base Sepolia and accepts payment in USDC (asset 0x036CbD53842c5426634e7929541eC2318f3dCF7e) with a maximum cost of 0.1 USDC per request (100000 in 6-decimal units) and a timeout of up to 1200 seconds.
The endpoint uses a simple POST interface: callers send a JSON body with a single required `input` string field and receive a JSON response containing a `response` string field. Payment is handled via the X-PAYMENT header following the x402 exact-payment scheme. No API key or separate authentication is needed beyond the on-chain payment.
Questflow, the provider behind Tars, operates a broader platform focused on autonomous AI "Clones" that aggregate prediction markets and on-chain trading venues such as Polymarket, Pump.fun, OKX, and Opinion Labs. Tars appears to be one of the agents within this ecosystem. Note that this endpoint runs on Base Sepolia (a testnet), so it is likely a development or staging deployment rather than a production service. No formal API documentation, OpenAPI spec, or detailed usage guides were found beyond the x402 challenge metadata.
Capabilities
Use cases
- —Breaking down complex multi-step requests into sub-tasks for specialized agents
- —Coordinating autonomous agent workflows via a single natural-language prompt
- —Pay-per-use AI planning without API keys, using on-chain USDC micropayments
- —Integrating an AI task planner into agent-to-agent pipelines via x402
Fit
Best for
- —Developers experimenting with x402 payment-gated AI endpoints
- —Agent-to-agent orchestration where a planner decomposes tasks
- —Quick prototyping of multi-step AI workflows on a testnet
Not for
- —Production workloads requiring mainnet settlement and SLAs
- —Use cases needing deterministic, auditable task execution with guaranteed outputs
- —Users who need detailed API documentation or OpenAPI specs before integrating
Quick start
curl -X POST https://api-dev.intra-tls2.dctx.link/x402/copilot/qrn:copilot:tars \
-H "Content-Type: application/json" \
-H "X-PAYMENT: <x402-payment-token>" \
-d '{"input": "Plan a strategy to launch a token on Base"}'Example
Request
{
"input": "Break down the steps to create and deploy an ERC-20 token on Base, including auditing and liquidity setup."
}Response
{
"response": "Here is a plan broken into steps: 1) Draft the ERC-20 smart contract with OpenZeppelin templates. 2) Write and run unit tests using Foundry. 3) Submit the contract for a third-party audit. 4) Deploy to Base mainnet via a multisig. 5) Seed initial liquidity on Uniswap V3 Base pool. 6) Announce and verify the contract on Basescan."
}Endpoint
Quality
The x402 challenge is live and provides a clear input/output schema, but there is no OpenAPI spec, no formal documentation, and the endpoint runs on a testnet (Base Sepolia). The response format is inferred from the outputSchema in the 402 challenge; no actual successful response was captured.
Warnings
- —Endpoint runs on Base Sepolia (testnet) — likely a development/staging deployment, not production.
- —No OpenAPI spec or formal API documentation found.
- —Example response is inferred from the outputSchema, not from an actual observed response.
- —Maximum timeout of 1200 seconds (20 minutes) suggests potentially long-running operations.
Citations
- —Endpoint returns HTTP 402 with x402 version 1 challenge requiring USDC payment on Base Sepoliahttps://api-dev.intra-tls2.dctx.link/x402/copilot/qrn:copilot:tars
- —maxAmountRequired is 100000 (0.1 USDC) with maxTimeoutSeconds of 1200https://api-dev.intra-tls2.dctx.link/x402/copilot/qrn:copilot:tars
- —Questflow operates an AI Clone platform aggregating Polymarket, Pump.fun, OKX, and Opinion Labshttps://api-dev.intra-tls2.dctx.link
- —Tars is described as an AI task planner and coordinator that breaks down complex requests and delegates to specialized agentshttps://api-dev.intra-tls2.dctx.link/x402/copilot/qrn:copilot:tars