AgentShare MCP Registry: Discover and List MCP Servers via agent.json and

AgentShare MCP Registry provides a curated, machine-readable directory via agent.json for Claude Code users. Agents can submit listings using x402 micropayments and track analytics.

Key Takeaways

• AgentShare MCP Registry provides a curated, machine-readable directory via agent.json for Claude Code users.
• Agents can submit listings using x402 micropayments and track analytics.

What Changed — AgentShare MCP Registry Launches with Curated Listings and Agent-Native Submission

AgentShare MCP Registry is a new, curated directory for Model Context Protocol (MCP) servers, designed for both humans and autonomous AI agents. It launched with features that solve the discovery problem: great MCP servers hidden in GitHub repos or buried READMEs. The registry offers machine-readable discovery via agent.json, human-reviewed listings, owner analytics, and agent-native submission through x402 micropayments.

What It Means For You — Concrete Impact on Daily Claude Code Usage

If you build or use MCP servers with Claude Code, this registry changes how you find and distribute them. Instead of scraping GitHub or relying on word-of-mouth, you can:

• Discover MCP servers programmatically: Use curl -s https://agentshare.dev/agent.json | jq '.mcp_listings' to get a machine-readable list of verified servers. This is ideal for agents that need to dynamically select tools.
• List your own MCP server: Submit for free (human review takes 24–48 hours) and get a canonical profile page at https://agentshare.dev/registry/{id}. You can also track analytics like opens, referrers, and countries.
• Submit via agent: Agents can pay a small USDC micropayment to mint a scoped registry:submit key, then submit metadata via API. This enables fully autonomous listing workflows.
• Get smart tag suggestions: Use POST /api/v1/registry/suggest-tags to auto-generate relevant tags from your MCP server's name, description, and GitHub URL.

Try It Now — Commands, Config, or Prompts to Take Advantage of This

Discover MCP Servers

# List all verified MCP servers
curl -s https://agentshare.dev/agent.json | jq '.mcp_listings'

# Get compact summary
curl -s https://agentshare.dev/agent.json | jq '.summary_json'

List Your MCP Server (Human)

1. Go to agentshare.dev/registry
2. Submit your MCP server details (name, description, GitHub URL, MCP endpoint)
3. Wait 24–48 hours for human review
4. Once approved, get your canonical page and analytics dashboard

List Your MCP Server (Agent via x402)

# Step 1: Mint a scoped key with micropayment
POST https://agentshare.dev/api/v1/registry/agent-key

# Step 2: Submit listing
POST https://agentshare.dev/api/v1/registry/submit
X-API-Key: agshp_...scoped...

Suggest Tags for Your MCP

curl -X POST https://agentshare.dev/api/v1/registry/suggest-tags \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Fashion Inventory MCP",
    "description": "Track apparel SKUs and size variants for retail merchandising bots.",
    "category": "integrations",
    "github_url": "https://github.com/you/fashion-mcp"
  }'

Why This Matters for Claude Code Users

Claude Code relies on MCP servers to extend its capabilities—accessing databases, APIs, or custom tools. The AgentShare registry provides a trusted, machine-readable source for discovering these servers. Unlike generic directories, it offers curated review, analytics, and agent-native submission, aligning with the trend toward MCP minimalism (fewer, higher-quality servers) and the stateless MCP specification (removing sessions).

If you ship an MCP server, listing it here ensures visibility for both human developers and autonomous agents. If you consume MCP servers, use the agent.json endpoint to dynamically fetch verified tools without scraping HTML.

Architecture Notes for Builders

• Async analytics keep redirects fast
• Privacy: stores ip_hash (SHA-256), not raw IPs
• Dual auth: API key or x402 micropayments
• Scoped keys: registry:submit keys cannot access other AgentShare APIs

FAQ

Is listing free? Yes for humans and agents. x402 on agent-key mint is a small anti-spam layer.

How long does review take? Typically 24–48 hours. Save your status_url after submit.

Can agents list without a human account? Yes—mint a scoped key via x402, submit via API, poll status with the submission token.

What analytics do owners get? Views/opens over time, top referrers, user-agent labels, country breakdown (aggregated).

---

Source: dev.to

---

Originally published on gentic.news