MCP server and A2A agent for AgentsPodium hosting: create, watch, pay for and connect hosted AI agent pods
- ✓Actively maintained (<30d)
- ✓Clear description
- ✓Documented (README)
- !No standard license detected
git clone https://github.com/agentspodium/agent-gateway{
"mcpServers": {
"agent-gateway": {
"command": "node",
"args": ["/path/to/agent-gateway/dist/index.js"]
}
}
}MCP Servers overview
# agent-gateway
Lets other AI agents drive AgentsPodium hosting (create/monitor/manage
instances) through two machine protocols — MCP (Streamable HTTP) and A2A
(Agent2Agent JSON-RPC) — both backed by the existing account API at
`https://agentspodium.com/api`.
The gateway holds **no state and no secrets of its own**. Every call carries
the caller's own AgentsPodium API key (`Authorization: Bearer ak_live_...`, or
a session token), which is forwarded verbatim to the account API. Missing
that header always answers `401` with a JSON body pointing at
https://agentspodium.com/account ("API keys for agents") and
https://hosting.defispace.com/docs/auth.
One process serves both hostnames, selected by the `Host` header wherever a
path is shared between the two (only `GET /`); everything else defaults to
MCP behaviour, and `mcp.agentspodium.com` / `a2a.agentspodium.com` route to
the same Deployment.
## Run locally
```bash
npm install
npm run build
ACCOUNT_API=https://agentspodium.com/api PORT=3200 npm start
# or, for iteration:
npm run dev
```
Env vars:
- `ACCOUNT_API` — base URL of the account API (default `https://agentspodium.com/api`).
- `PORT` — HTTP port (default `3200`).
- `A2A_HOST` — hostname that selects A2A behaviour for `GET /` (default `a2a.agentspodium.com`).
## Calling it
### MCP — `mcp.agentspodium.com/mcp`
Streamable HTTP, stateless (no session id). Requires
`Accept: application/json, text/event-stream`.
```bash
# initialize
curl -s -X POST https://mcp.agentspodium.com/mcp \
-H "Authorization: Bearer ak_live_..." \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{
"jsonrpc": "2.0", "id": 1, "method": "initialize",
"params": {
"protocolVersion": "2025-06-18",
"capabilities": {},
"clientInfo": { "name": "example-client", "version": "0.1.0" }
}
}'
# call a tool
curl -s -X POST https://mcp.agentspodium.com/mcp \
-H "Authorization: Bearer ak_live_..." \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{
"jsonrpc": "2.0", "id": 2, "method": "tools/call",
"params": { "name": "list_platforms", "arguments": {} }
}'
```
Tools: `list_platforms`, `list_instances`, `create_instance`,
`get_instance_health`, `get_instance_term`, `set_llm_key`, `pause_instance`,
`resume_instance`, `rebuild_instance`, `delete_instance`, `get_payment_options`,
`set_peers`. Plus one resource, `agentspodium://docs` (the hosting docs).
### A2A — `a2a.agentspodium.com/`
```bash
curl -s https://a2a.agentspodium.com/.well-known/agent-card.json
# structured (data part)
curl -s -X POST https://a2a.agentspodium.com/ \
-H "Authorization: Bearer ak_live_..." \
-H "Content-Type: application/json" \
-d '{
"jsonrpc": "2.0", "id": 1, "method": "message/send",
"params": {
"message": {
"parts": [{ "kind": "data", "data": { "skill": "create-instance", "params": { "engine": "hermes", "tier": "small", "name": "My Agent" } } }]
}
}
}'
# plain text (small command grammar)
curl -s -X POST https://a2a.agentspodium.com/ \
-H "Authorization: Bearer ak_live_..." \
-H "Content-Type: application/json" \
-d '{
"jsonrpc": "2.0", "id": 2, "method": "message/send",
"params": { "message": { "parts": [{ "kind": "text", "text": "platforms" }] } }
}'
```
Text grammar: `create <engine> <tier> [name ...]`, `health <id>`,
`term <id>`, `platforms`, `payment <id>`. Anything else answers with the
skill list and the exact JSON shape.
Public, no auth: `GET /catalog.json` (proxy of the account API's
`/a2a-catalog`).
## Deploy
```bash
docker build -t agentpodium-agent-gateway apps/agent-gateway
# push, then apply:
kubectl apply -f apps/agent-gateway/deploy/gateway.yaml
```
`deploy/gateway.yaml` creates namespace `ap-agentgw`, a Deployment/Service on
port 3200, and two `HTTPRoute`s (`mcp.agentspodium.com`,
`a2a.agentspodium.com`) against the shared `akash-gateway` Gateway. Swap
`__IMAGE_TAG__` for the built image tag before applying.
What people ask about agent-gateway
What is agentspodium/agent-gateway?
+
agentspodium/agent-gateway is mcp servers for the Claude AI ecosystem. MCP server and A2A agent for AgentsPodium hosting: create, watch, pay for and connect hosted AI agent pods It has 0 GitHub stars and its last recorded update is dated 2026-09-08.
How do I install agent-gateway?
+
You can install agent-gateway by cloning the repository (https://github.com/agentspodium/agent-gateway) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is agentspodium/agent-gateway safe to use?
+
Our security agent has analyzed agentspodium/agent-gateway and assigned a Trust Score of 62/100 (tier: OK). See the full breakdown of passed checks and flags on this page.
Who maintains agentspodium/agent-gateway?
+
agentspodium/agent-gateway is maintained by agentspodium. The last recorded GitHub activity is dated 2026-09-08, with 0 open issues.
Are there alternatives to agent-gateway?
+
Yes. On ClaudeWave you can browse similar mcp servers at /categories/mcp, sorted by popularity or recent activity.
Deploy agent-gateway to your cloud
Ship this repo to production in minutes. Each platform spins up its own environment with editable env vars.
Maintain this repo? Add a badge to your README
Drop the badge into your GitHub README to show it's tracked on ClaudeWave. Each badge links back to this page and reflects the live Trust Score.
[](https://claudewave.com/repo/agentspodium-agent-gateway)<a href="https://claudewave.com/repo/agentspodium-agent-gateway"><img src="https://claudewave.com/api/badge/agentspodium-agent-gateway" alt="Featured on ClaudeWave: agentspodium/agent-gateway" width="320" height="64" /></a>More MCP Servers
Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
An open-source AI agent that brings the power of Gemini directly into your terminal.
Real-time global intelligence dashboard. AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface
The fastest path to AI-powered full stack observability, even for lean teams.
🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl!