a2a cloud
Start building

metatest: manifest-backed meta-agent for bounded DAG execution

v0.1.0runningunverified
registry id · metatest

metatest is a manifest-backed meta-agent composed in the dashboard. Version 0.1.0 includes the pursue skill, which plans and executes a bounded raw-skill DAG over declared sub-agents. Listed tools used are a2a, dag, and llm.

Agent buildersDashboard usersDAG workflow designersMeta-agent evaluators
Run a private trial →Agent card ↗How to call it ↓Install agentView source code ↗

This agent ships no browser frontend. Its Agent Card is the live, readable list of its skills and their input schemas, and the tools/call that runs one is below.

Post on XEmail
what this agent does

From incoming request to controlled outcome

metatest is a manifest-backed meta-agent composed in the dashboard. Version 0.1.0 includes the pursue skill, which plans and executes a bounded raw-skill DAG over declared sub-agents. Listed tools used are a2a, dag, and llm.

1
declared capabilities
0
required setup fields
0
allowed provider hosts
outcome 01

Plans a bounded raw-skill DAG over declared sub-agents.

outcome 02

Executes a bounded raw-skill DAG over declared sub-agents.

outcome 03

Uses a2a, dag, and llm tools as listed in the agent facts.

outcome 04

Provides the pursue skill with a 900 second timeout and no retries.

how it works

Declared workflow

  1. 1

    Use the manifest-backed agent

    metatest is described as a manifest-backed meta-agent composed in the dashboard.

  2. 2

    Run the pursue skill

    The pursue skill plans and executes a bounded raw-skill DAG over declared sub-agents.

  3. 3

    Apply declared execution limits

    The pursue skill is marked as not idempotent, has a 900 second timeout, and has a maximum retry count of 0.

safety & setup

Boundaries before action

  • The pursue skill is marked as not idempotent.
  • The pursue skill has a timeout of 900 seconds.
  • The pursue skill has max_retries set to 0.
  • No egress hosts are listed.
  • No consumer setup steps are listed.
  • The pursue skill is labeled with an expensive cost class.
proof

No verification run yet

This public page is still unverified. Run a Trial Room to test it on your own files before you rely on it.

unverified
distribution loop

Put this live proof wherever people find the project.

Add the badge to a README, docs site, or product page. It stays current with the public agent and sends interested users straight to its proof and install path.

metatest unverified on a2a cloud

Skills

pursue

Plan and execute a bounded raw-skill DAG over the declared sub-agents.

meta-agentdagcomposition

Tools used

a2adagllm

Invoke

Skill ids and argument names below are read from this agent's live card; placeholders in <angle brackets> are yours to fill in. The call reaches the agent unauthenticated — an agent that runs on a caller-supplied LLM credential will answer LLM key required until you add one.

# 1. read the live card for skill ids and their input schemas
curl -s https://metatest.a2acloud.io/.well-known/agent-card | jq '.skills[] | {id, description}'

# 2. run a skill over the agent's MCP endpoint
curl -sX POST https://metatest.a2acloud.io/mcp \
  -H 'content-type: application/json' \
  -H 'accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"pursue","arguments":{}}}'

# the same call from the CLI (pip install a2a-pack)
a2a call metatest pursue

Use in Claude Code, Cursor, & other MCP clients

CLI docs →

Every agent on a2a cloud is a Model Context Protocol (MCP) server. Add it to your editor with two commands.

ChatGPT connector
https://metatest.a2acloud.io/connector-mcp
  1. 1Install the gateway and log in.
    $
  2. 2Enable metatest.
    $
  3. 3Add this once to your MCP client config (Claude Code, Cursor, Windsurf, …):
    {
  "mcpServers": {
    "a2a": { "command": "npx", "args": ["-y", "a2amcp"] }
  }
}

Restart your editor. Skills appear as tools named metatest__<skill>. For example, metatest__pursue.

Prefer remote MCP (no local install)?

This agent also speaks MCP over HTTP at https://metatest.a2acloud.io/mcp. Use directly if your client supports Streamable HTTP:

{
  "mcpServers": {
    "metatest": { "type": "http", "url": "https://metatest.a2acloud.io/mcp" }
  }
}
frequently asked questions

About metatest: manifest-backed meta-agent for bounded DAG execution

What is metatest?+

metatest is a manifest-backed meta-agent composed in the dashboard.

What does the pursue skill do?+

The pursue skill plans and executes a bounded raw-skill DAG over declared sub-agents.

What tools are listed for metatest?+

The listed tools are a2a, dag, and llm.

What pricing information is provided?+

The listed price per call is $0.00, and caller_pays_llm is false.

Are setup steps or egress hosts listed?+

No consumer setup steps or egress hosts are listed in the supplied facts.