Dataset · Hugging Face

springofwindslabs/mcp-agent-trajectory-benchmark

Name: mcp-agent-trajectory-benchmark
Creator: springofwindslabs
License: apache-2.0
Keywords: task_categories:text-generation, language:en, license:apache-2.0, size_categories:n<1K, format:json, modality:text, library:datasets, library:pandas, library:polars, library:mlcroissant, region:us, mcp, function-calling, tool-use, agents

🛑 Stop LLM Agent Collapse Caused by State Corruption Most agent failures are not tool failures. They are state failures. The model believes the world is still valid — when reality has already changed. This MCP trajectory dataset trains belief revision, state recovery, and autonomous replanning: believed : 45/50 rooms synced ← agent trusts a stale worker log tool : "success", count: 0 ← silent no-op (the worst failure mode) verify : GDS reports actual = 40 ←… See the full description on the dataset page: https://huggingface.co/datasets/springofwindslabs/mcp-agent-trajectory-benchmark.

View on Hugging FaceReport an issue

Author / organization	springofwindslabs
License (as declared)	apache-2.0
Downloads	317 (as reported by Hugging Face)
Likes	0
Created	17 September 2026
Last updated at source	8 October 2026

mcpfunction-callingtool-useagentssyntheticerror-recovery

Embed this entry

<iframe src="https://silytics.si/embed/item/c02d45ae-f6aa-49ff-94f6-61f80248b29a" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Popularity, last 2 days

317 now (-26 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.

More from springofwindslabs

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.