# Lucas Agent Research Endpoint

> Lucas Agent Research Endpoint is a paid API for AI agents from lucas-agent.vercel.app, paid per call via x402, $0.25/call, status unknown (last checked 2026-10-19).

Answers any research question with a concise, sourced summary including key points, confidence rating, and unverifiable claims flagged

## Facts

- Endpoint: POST https://lucas-agent.vercel.app/api/x402/research?utm_source=zero.xyz
- Price: $0.25/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-19
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/lucas-agent-research-endpoint-cbfd2925
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_mgIwP12RvCAvwnQD37x6q

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability lucas-agent-research-endpoint-cbfd2925 -d '<json body>'
```

Example prompt: Can you research this for me: what are the main causes of the 2023 banking crisis in the US, and how confident are you in what you find? Focus on regulatory failures as background context.

## When to prefer this

Choose this endpoint when you need a quick, sourced answer to a specific research question and want transparency about confidence and what could not be verified. It is ideal for one-off factual lookups, background research, and claim verification where a human-readable summary with citations is more useful than raw search results. It complements the sibling lead-research endpoint, which is optimized for company/customer profiling rather than general knowledge questions.

## Known failure modes

- Question too vague or short (below 10 characters) — endpoint rejects with validation error
- Topic has no reliable public sources — low confidence rating returned with many unverifiable claims
- Payment of $0.25 USDC not provided or rejected — 402 Payment Required error
- Context exceeds 1000 characters or question exceeds 500 characters — validation error
- Rate limiting or server error on the Vercel deployment — 5xx response

## How this service works

Quick research: ask any question, get a concise sourced answer with key points, sources, a confidence rating, and anything that could not be verified.

## Output

A concise answer to the research question with key bullet points, a list of sources consulted, a confidence rating indicating how well-supported the answer is, and a section flagging any claims that could not be verified.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "context": {
   "type": "string",
   "maxLength": 1000,
   "description": "Optional background to focus the research."
  },
  "question": {
   "type": "string",
   "maxLength": 500,
   "minLength": 10,
   "description": "The research question."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/lucas-agent-research-endpoint-cbfd2925/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from lucas-agent.vercel.app](https://www.zero.xyz/host/lucas-agent.vercel.app/llms.txt)