
Nlp
Failed Tool Calls (JSONL)
Failed tool invocation rows for agent error-handling and retry tests.
- File
- JSONL · Nlp · 4 records
- Use case
- Tool calling / function calling / Error handling
JSONL traces of function/tool calls with arguments and results, for testing agent harnesses and tool routers.
Each card starts with one representative test file. Choose a use case, search, or filter to find a specific profile; a card’s variant menu contains its related samples.
10 test families · 14 available files · Page 1 of 1

Nlp
Failed tool invocation rows for agent error-handling and retry tests.

Nlp
Multi-step agent action/observation trace for ReAct-style eval harnesses.

Nlp
OpenAI function tools definition JSON paired with the tool-calling traces.

Nlp
Parallel multi-tool call payloads for concurrent tool-router tests.

Nlp
Function-call result rows decoupled from chat messages for router testing.

Eval
SAMPLE tool-calling JSON (tool-call-search) for agent harness schema tests.


Nlp
Minimal single-turn tool-calling trace for quick parser smoke tests.

Nlp
Anthropic-style tool_use blocks with paired tool results for agent harness tests.

Nlp
OpenAI-style tool-call JSONL with function invocation and tool result messages.