hive-rank about · visibility · docs · pricing · sign in · sign up free

Sell to AI agents.

Claude Code decides what your customers install. Hive Rank watches the retrieval surfaces agents read and tells you when you win the pick — and who took it when you don't.

Free to read · contributions mint credits · no card required.

# install — wires the MCP into Claude Code, OpenCode, Codex
$ npx hive-rank

→ 23 skills installed
→ MCP server: mcp.hive-rank.com
→ capture hooks enabled

# then run any hive skill inside your agent
$ claude "/hive:trends"

Agents don't browse. They decide.

The agent purchase funnel has four steps and no ad slots. A human at #4 might still click you. An agent takes the top answer and moves on.

Retrieved — your domain came back in the agent's search. Not retrieved → invisible to agents.
Read — the agent fetched your docs, pricing, README. Unparseable pages end the deal here.
Picked — the agent chose one option and acted on it. Second place is a loss, and we tell you who took it.
Installed — npx your-package. No landing page, no demo call. The deal closes in the terminal.

What this is

In-session observations. Every WebSearch and WebFetch Claude Code makes in a hive session is anonymized and contributed — no scraping, no schedule. You see what the agent saw.
Hive Score (0–100). Position alone lies. The score blends visibility, breadth, consensus, and freshness so you know whether a #1 is load-bearing or a fluke.
23 slash commands, one MCP. Rankings, score, trending, schema, audits, network status, monitors. see all 23 →
Frameworks. Auto-wired: Claude Code, OpenCode, Codex. Manual: Cursor, Continue, Aider, or any MCP client.

	SEO tools	Hive Rank
sampling	scheduled crawls (Ahrefs, Semrush)	real agent searches, as they happen
unit	keyword → ranking page	task → picked domain
freshness	daily/weekly refresh	live network
price	$100–500/mo	free to read · contribute to earn

Install

One command, any shell. The installer detects Claude Code, OpenCode, and Codex:

$ npx hive-rank

Any other MCP client:

{
  "mcpServers": {
    "hive-rank": { "url": "https://mcp.hive-rank.com/mcp" }
  }
}

Contribute to earn

/hive:<skill>   → activates the session
web searches   → captured + anonymized
the hive       → aggregates + normalizes
you (authed)   → credits mint to your ledger
privacy        → SHA-256 identity hash, no emails

FAQ

Is this AI SEO, GEO, or LLM visibility?

Related, but different — we call the category agent visibility. Those tools chase answer engines and the chat surface. Hive Rank measures how coding agents discover and pick packages, docs, and domains — currently capturing Claude Code's WebSearch/WebFetch traffic, with OpenCode + Codex hooks next.

Which agents does the MCP support?

The installer auto-wires Claude Code, OpenCode, and Codex CLI. The MCP speaks the open protocol — Cursor, Windsurf, Continue, and any custom client connect with the JSON config above.

What data do you actually collect?

From the contribution hook: queries, URLs, and observed positions — public data the agent already fetched. Your contributor identity is a one-way SHA-256 hash; no email, no account needed to contribute. The dashboard does use email + password so you can manage tokens, monitors, and credits — the two paths are separate. See the docs for the full schema.

How is this free?

Contribute-to-earn. Invoke a /hive:* skill and the hook captures the web searches the agent makes next in that session as anonymized signal. In return you read everyone else's contributions. Hatchling tier hands you 500 credits on signup; authenticated contributions mint more.

Can I opt out of contributing?

Run npx hive-rank --uninstall to remove the hooks. To purge data already sent, email privacy@hive-rank.com with the first 8 chars of your contributorId (run /hive:privacy to see it). An in-product toggle lands with the read-only paid tier.

Your next thousand customers are agents.

We tell you whether they're buying you or your competitor. sign up free →