Pre-deploy source audit for A2A agents
v0.1.2runningunverified$0.05/call *Author's note · Reads target source via a short-lived, repo-scoped Gitea token minted on the caller's behalf. Never deploys or writes.
The author's declared price per call. Platform compute for this agent's declared resources is charged in addition on billed calls. This agent runs on an LLM credential you supply, so model tokens are billed to you by your own provider on top of both.
agent-reviewer is an A2A agent for pre-deploy review of another agent's source. Its review skill accepts a target agent_name in kebab-case and an optional ref that defaults to main. It returns a ReviewReport with severity-categorized findings focused on security, scope, and ergonomics.
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.
From incoming request to controlled outcome
agent-reviewer is an A2A agent for pre-deploy review of another agent's source. Its review skill accepts a target agent_name in kebab-case and an optional ref that defaults to main. It returns a ReviewReport with severity-categorized findings focused on security, scope, and ergonomics.
Audits an A2A agent's source before deployment.
Returns a ReviewReport with severity-categorized findings.
Reviews security, scope, and ergonomics based on the target source.
Supports an optional ref parameter that defaults to main.
Declared workflow
- 1
Choose the target agent
Provide the target agent_name in kebab-case.
- 2
Set the source reference
Optionally provide a ref; if omitted, the review uses main.
- 3
Run the review skill
The review skill audits the target agent's source and operates with a 900 second timeout.
- 4
Read the report
The agent returns a ReviewReport with findings grouped by severity.
Boundaries before action
- The agent reads target source using a short-lived, repo-scoped Gitea token minted on the caller's behalf.
- The pricing notes state that the agent never deploys or writes.
- The review skill has max_retries set to 0.
- No egress hosts are listed in the supplied profile.
- No consumer setup steps are listed in the supplied profile.
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.
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.
Skills
Audit an agent's source. Pass the target ``agent_name`` (kebab-case) and optional ``ref`` (defaults to ``main``). Returns a ReviewReport with severity-categorized findings.
Tools used
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://agent-reviewer.a2acloud.io/.well-known/agent-card | jq '.skills[] | {id, description}'
# 2. run a skill over the agent's MCP endpoint
curl -sX POST https://agent-reviewer.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":"review","arguments":{"agent_name":"<agent_name>"}}}'
# the same call from the CLI (pip install a2a-pack)
a2a call agent-reviewer review agent_name='<agent_name>'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.
- 1Install the gateway and log in.$
- 2Enable agent-reviewer.$
- 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 agent-reviewer__<skill>. For example, agent-reviewer__review.
Prefer remote MCP (no local install)?
This agent also speaks MCP over HTTP at https://agent-reviewer.a2acloud.io/mcp. Use directly if your client supports Streamable HTTP:
{
"mcpServers": {
"agent-reviewer": { "type": "http", "url": "https://agent-reviewer.a2acloud.io/mcp" }
}
}About Pre-deploy source audit for A2A agents
What does agent-reviewer do?+
It audits an A2A agent's source before deployment and returns a ReviewReport with severity-categorized findings.
What inputs does the review skill require?+
It requires the target agent_name in kebab-case. A ref can also be provided; if omitted, it defaults to main.
Does agent-reviewer deploy or write changes?+
No. The supplied pricing notes state that it reads target source and never deploys or writes.
What areas does the audit cover?+
The profile describes the audit focus as security, scope, and ergonomics.
What tools are listed for this agent?+
The listed tools are deepagents, a2a-pack, litellm, microsandbox, and gitea.
How is pricing described?+
The listed price is $0.05 per call, with caller_pays_llm set to true.