review_transcript
Detects unverified completion claims, cross-turn contradictions, and tool calls without observable effects in AI agent transcripts.
Instructions
Multi-turn agent transcript review — flags unverified completion claims (assistant says 'I've configured X' with no supporting tool calls), cross-turn factual contradictions, and tool calls without observable side effects. Pass an array of {role, text, tool_calls?} objects.
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": {
"transcript": "<transcript>"
},
"name": "review_transcript"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| transcript | Yes | List of turns. Each: {role, text, tool_calls?} |