MCPMark Trajectory Logs (mcpmark-v1-0905)
0
10 commits
1 linked in READMEs
updated Sep 9, 2025
These logs are publicly available for research use. They capture end-to-end trajectories (prompts, tool calls, and outcomes) for MCPMark tasks across multiple MCP services and models.
Each task run produces a trajectory folder containing three files:
meta.json: Metadata (task id, model, timestamps, status, etc.)messages.json: Turn-by-turn exchanges including model thoughts/tool callsexecution.log: Key execution-time logsTypical structure:
mcpmark-v1-0905/
<model>__<mcp_service>/
run-1/
<category>__<task>/
meta.json
messages.json
execution.log
For fast browsing and inspection, use the MCPMark Trajectory Explorer. You can drag-and-drop either a single messages.json or a full trajectory folder (the three files above):
10 commits
MCPMark Trajectory Logs (mcpmark-v1-0905)
0
10 commits
1 linked in READMEs
updated Sep 9, 2025
These logs are publicly available for research use. They capture end-to-end trajectories (prompts, tool calls, and outcomes) for MCPMark tasks across multiple MCP services and models.
Each task run produces a trajectory folder containing three files:
meta.json: Metadata (task id, model, timestamps, status, etc.)messages.json: Turn-by-turn exchanges including model thoughts/tool callsexecution.log: Key execution-time logsTypical structure:
mcpmark-v1-0905/
<model>__<mcp_service>/
run-1/
<category>__<task>/
meta.json
messages.json
execution.log
For fast browsing and inspection, use the MCPMark Trajectory Explorer. You can drag-and-drop either a single messages.json or a full trajectory folder (the three files above):
10 commits