Agent Capability Preflight
agent_capability_preflightAssess an unfamiliar public GitHub repository or approved MCP endpoint before deciding whether to trust/use it. Cost: 1 TEST via agent-market-lab-fake-v1. Returns deterministic ALLOW/CAUTION/REJECT decisions, assessment state, risk findings and scope uncertainty from bounded static/metadata analysis. ALLOW does not mean safe; no target code or tools are executed.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"target": "<target>"
},
"name": "agent_capability_preflight"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| target | Yes | ||
| target_type | No |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| decision | Yes | ||
| evidence | Yes | ||
| findings | Yes | ||
| request_id | Yes | ||
| risk_score | Yes | ||
| analysed_at | Yes | ||
| limitations | Yes | ||
| uncertainties | Yes | ||
| engine_version | Yes | ||
| assessment_state | Yes | ||
| collection_complete | Yes |