# Agent-Dojo SPAR Runner

> Agent-Dojo SPAR Runner is a paid API for AI agents from apiwitchcraft.duckdns.org, paid per call via x402, $0.002/call, status unknown (last checked 2026-09-14).

Runs a SPAR (agent scenario) in the Agent-Dojo framework, executing a defined agent logic against a chosen scenario for robustness testing.

## Facts

- Endpoint: POST https://apiwitchcraft.duckdns.org/agent-dojo/spar
- Price: $0.002/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-14
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agent-dojo-spar-runner-8ce3d9fa
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_qfla1GYdwxOJ8sFPG0dFh

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability agent-dojo-spar-runner-8ce3d9fa -d '<json body>'
```

Example prompt: Run a SPAR simulation in Agent-Dojo using scenario ID 'phishing-resistance-01' with my agent logic defined as the provided strategy object — I want to see how it performs.

## When to prefer this

Choose this endpoint when you need to evaluate AI agent behavior against structured benchmark scenarios using the Agent-Dojo SPAR framework, particularly for robustness and adversarial testing. Prefer it over generic simulation tools when you have a specific scenarioId and agent logic object ready.

## Known failure modes

- Invalid or unknown scenarioId returns an error
- Malformed agentLogic object causes a schema validation failure
- Payment failure (x402) if USDC balance is insufficient
- Scenario execution timeout if agent logic produces infinite loops
- Service unavailable if the duckdns host is unreachable

## How this service works

AgentDojo SPAR endpoint: score your LLM agent's security against prompt-injection attacks. Submit an agent trace and get SPAR (Security, Privacy, Attack-Resistance) scores plus pass/fail on injected-goal detection. x402 pay-per-call on Base USDC. For red-teamers, AI-safety labs, and agents self-evaluating robustness to adversarial prompts.

## Output

Returns the outcome of the agent scenario run, including performance metrics, agent decisions taken during the scenario, and any robustness or failure signals produced by the SPAR framework.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "agentLogic": {
   "type": "object"
  },
  "scenarioId": {
   "type": "string"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agent-dojo-spar-runner-8ce3d9fa/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from apiwitchcraft.duckdns.org](https://www.zero.xyz/host/apiwitchcraft.duckdns.org/llms.txt)