# secant-agent-research-pack

secant-agent-research-pack is an MCP server for coding agents to search the web, read pages, create cited briefs, and track changes, with per-call billing.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 165 installs
- Author: hello-j2mo
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/hello-j2mo/secant-agent-research-pack
- Updated: 2026-05-25
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/secant-agent-research-pack

## What it does

Five tools sit behind one remote MCP endpoint. `search` runs web queries, `extract_page` reads between one and twenty URLs per call, and `research_pack` does a query plus page reads plus citations in a single request.

\`monitor_diff` watches a page and reports what moved since the last look. `codex_audit` is the entrypoint for Codex audits: jobs are queued through the REST `/codex/jobs` endpoint. JSON comes back normalized, sources attached.

Calls are paid. Settlement runs through x402 in USDC on Base, so payment has to be sorted out before an agent gets an answer — a call without it does not go through. The endpoint is the same for every client: `https://secant-agent-research-pack--hello-j2mo.run.tools`.

## Requirements
- Remote server: `https://secant-agent-research-pack--hello-j2mo.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http secant-agent-research-pack https://secant-agent-research-pack--hello-j2mo.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add secant-agent-research-pack --url https://secant-agent-research-pack--hello-j2mo.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "secant-agent-research-pack": {
            "url": "https://secant-agent-research-pack--hello-j2mo.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "secant-agent-research-pack": {
            "type": "http",
            "url": "https://secant-agent-research-pack--hello-j2mo.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http secant-agent-research-pack https://secant-agent-research-pack--hello-j2mo.run.tools
```

## When to use it
- pull source links on a topic without losing track of them
- read a batch of pages in one call and get clean text back
- hand the agent a cited brief instead of raw snippets
- catch edits on a page between two runs
- queue a Codex audit job and pick up the result

## Example prompts
- Run `search` for recent write-ups on x402 and keep every source URL.
- Use `extract_page` on these eight URLs and give me back normalized JSON.
- Point `monitor_diff` at the pricing page and tell me what changed.

## Similar
- [deep-research](https://affpapa.org/en/nevibecode/skills/affaan-m-deep-research.md) — skill · 7000 installs
- [tavily-research](https://affpapa.org/en/nevibecode/skills/tavily-research.md) — skill · 25100 installs
- [research](https://affpapa.org/en/nevibecode/skills/alireza-rezvani-research.md) — skill · 348 installs
- [WebberSites x402 Data API](https://affpapa.org/en/nevibecode/mcp/webbersites-x402-data-api.md) — MCP server · 832 installs
- [AgentScrape](https://affpapa.org/en/nevibecode/mcp/agentscrape.md) — MCP server · 175 installs
- [fetcher](https://affpapa.org/en/nevibecode/skills/fetcher.md) — skill · 4700 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
