Mock Agent

A2A agent server for mocking and testing. Uses a mock LLM client - no API keys required!

A enterprise-ready Agent-to-Agent (A2A) server that provides AI-powered capabilities through a standardized protocol.

Quick Start

The generated binary is a CLI. start boots the A2A server; --help and --version work as you'd expect.

# Run the agent
go run . start

# Or build and invoke the CLI directly
task build
./bin/mock-agent start

# Or with Docker
docker build -t mock-agent .
docker run -p 8080:8080 mock-agent

CLI

Command	Description
mock-agent start	Start the A2A server (blocks until SIGINT/SIGTERM)
mock-agent --help	Show top-level help (and per-subcommand with <cmd> --help)
mock-agent --version	Print the embedded version and exit

Quick Install

Add this agent to your Inference Gateway CLI:

infer agents add mock-agent http://localhost:8080 \
  --oci ghcr.io/inference-gateway/mock-agent:latest \
  --run

Features

• ✅ A2A protocol compliant
• ✅ AI-powered capabilities
• ✅ Streaming support
• ✅ OpenTelemetry instrumentation
• ✅ Enterprise-ready
• ✅ Minimal dependencies

Endpoints

• GET /.well-known/agent-card.json - Agent metadata and capabilities
• GET /health - Health check endpoint
• POST /a2a - A2A protocol endpoint

Available Tools

Tool	Description	Parameters
Read	Read a file from disk. Returns its contents, optionally sliced by line offset/limit. Use this to load SKILL.md bodies on demand.	file_path, offset, limit
echo	Echo back the input message (useful for basic connectivity tests)	message
delay	Simulate slow responses with configurable delays	duration_seconds, message
error	Simulate error conditions for testing error handling	error_type, message
random_data	Generate random test data	count, data_type
validate	Validate input against common patterns	input, validation_type
simulate_tool_call	Simulate a single tool call for load, latency and failure testing. Emits an instrumented span (gen_ai.tool.name) with a configurable duration and optional error status. The mock LLM drives this once per entry of a multi-tool-call workload.	duration_ms, fail, name

Examples

Example	Description
Connectivity smoke test	Send a known payload through the echo tool (the connectivity-check skill) and confirm the round-trip, verifying the agent is reachable and responding over A2A.
Error-handling drill	Drive the error tool across every error_type - validation, timeout, internal, not_found - via the error-injection skill to see how your client reacts to each failure mode.
Latency and load simulation	Combine the delay tool with random_data (the load-simulation skill) to return realistic test payloads after a configurable delay, exercising client timeouts and streaming under slow responses.
Distributed-tracing demo with nested tool spans	Send "read " to route through the Read built-in and emit a tool.read span nested under a2a.request, producing real per-tool sub-spans for end-to-end OpenTelemetry tracing (see examples/opentelemetry).

Skills (loaded into the system prompt)

Skill	Description	Source
connectivity-check	Use this when the user wants to verify the agent is reachable and responding correctly. Invokes the echo tool with a known payload and confirms the round-trip succeeded.	bare scaffold (.agents/skills/connectivity-check/SKILL.md)
error-injection	Use this when the user wants to test how their client handles different failure modes. Invokes the error tool across the supported error_type values (validation, timeout, internal, not_found) so the caller can observe each error path.	bare scaffold (.agents/skills/error-injection/SKILL.md)
load-simulation	Use this when the user wants to test client behavior under slow responses with realistic payloads. Combines the delay tool (to introduce latency) with the random_data tool (to produce a test payload of the requested shape).	bare scaffold (.agents/skills/load-simulation/SKILL.md)

Documentation

• Getting Started
• Configuration
• Usage
• Simulating Tool Calls

Configuration

The agent is configured via environment variables. Defaults are derived from agent.yaml; see CONFIGURATIONS.md for the full reference of custom and A2A_* variables.

Development

# Generate code from ADL
task generate

# Run tests
task test

# Build the application
task build

# Run linter
task lint

# Format code
task fmt

Adding Dependencies

The generator owns the baseline toolchain pins (SDK, server framework, logging, CLI, sandbox utilities). To extend the project without forking the templates, declare extras in agent.yaml - every empty list below is rendered by adl init --defaults precisely so it's discoverable:

Where	Purpose	Example entry	Rendered into
spec.language.go.vendor.deps	Runtime Go modules	github.com/stretchr/testify@v1.10.0	go.mod require block
spec.language.go.vendor.devdeps	Executable dev tools (Go 1.24 tool directive)	golang.org/x/tools/cmd/stringer@v0.20.0	go.mod tool directive
spec.development.deps	Cross-cutting sandbox tools (not tied to one language)	kubectl@1.31.0, terraform@1.9.5, deno@2.1.4	Flox manifest.toml / devcontainer feature

Entries use the <package>@<version> form. Built-in pins always win on conflict; the generator prints a warning and skips the user entry when shadowing is attempted. After editing agent.yaml, re-run task generate to refresh the manifests.

Debugging

Use the A2A Debugger to test and debug your A2A agent during development. It provides a web interface for sending requests to your agent and inspecting responses, making it easier to troubleshoot issues and validate your implementation.

docker run --rm -it --network host ghcr.io/inference-gateway/a2a-debugger:latest --server-url http://localhost:8080 tasks submit "What are your skills?"

docker run --rm -it --network host ghcr.io/inference-gateway/a2a-debugger:latest --server-url http://localhost:8080 tasks list

docker run --rm -it --network host ghcr.io/inference-gateway/a2a-debugger:latest --server-url http://localhost:8080 tasks get <task ID>

Deployment

Docker

The Docker image can be built with custom version information using build arguments:

docker build \
  --build-arg VERSION=1.2.3 \
  --build-arg AGENT_NAME="My Custom Agent" \
  --build-arg AGENT_DESCRIPTION="Custom agent description" \
  -t mock-agent:1.2.3 .

Available Build Arguments:

• VERSION - Agent version (default: 0.4.3)
• AGENT_NAME - Agent name (default: mock-agent)
• AGENT_DESCRIPTION - Agent description (default: A2A agent server for mocking and testing. Uses a mock LLM client - no API keys required!)

These values are embedded into the binary at build time using linker flags, making them accessible at runtime without requiring environment variables.

License

Apache 2.0 License - see LICENSE file for details