license: cc-by-4.0
task_categories:
- question-answering
- text-generation
language:
- en
tags:
- legal
- benchmark
- agents
- mcp
- deterministic-evaluation
pretty_name: CounselBench-100
size_categories:
- n<1K
configs:
- config_name: default
data_files:
- split: test
path: data/tasks.jsonl
CounselBench-100
CounselBench-100 v3.2.5 is a synthetic legal-work benchmark with 100 authored matters across ten practice workflows. Every task has a natural employee request, a 97-asset evidence room, twelve portfolio decisions, 5–9 supported actions, 3–7 evidence holds, and a distinct deep multi-provider MCP trajectory.
The answer is not preclassified in the evidence. Each portfolio item requires an immutable identity join, an operative-authority and revision lookup, a current-state comparison, and an effective approval/owner-capacity check across Clio, Gmail, Drive, and Slack. The agent then commits native provider state and reads each changed record back.
Public release
- Runnable Harbor world: https://hub.harborframework.com/datasets/blobfishai/counselbench-100
- Dataset and assets: https://huggingface.co/datasets/SamuelChien821/counselbench-100
- Benchmark explorer: https://blobfish.ai/benchmarks/counselbench-100
- Builder and verifier: https://github.com/blobfishai/legal-agent-simulation/tree/master/benchmark/counselbench100
Included files
data/tasks.jsonl: prompt, provider-bound assets, 14 semantic milestones, gold output, and metadata.task_files/: 9,700 unique files across Markdown, TXT, EML, CSV, JSON, XML, HTML, PDF, and XLSX.world/: pinned offline MCP implementation and hidden deterministic verifier.trajectories/: 100 solvability traces; these are excluded from model ranking.reports/: exact-version build, qualification, and conformance evidence.SCORING.md: causal, branch, state, containment, and readback contract.
Measured v3.2.5 release gates
| Gate | Measured |
|---|---|
| Tasks | 100 |
| Employee execution authority | 100/100 high-level requests authorize normal task-scoped closeout |
| Agent-visible files | 9,700 unique; all nine native formats parse |
| Required evidence reads | 58–86 per task |
| Reference MCP calls | 69–97 per task |
| Raw tool sequences | 100 distinct; maximum pair match 0.807453 |
| Semantic action graphs | 100 distinct; maximum pair match 0.165517 |
| Prompt maximum 5-shingle Jaccard | 0.267974 |
| Oracle and deterministic replay | 100/100 each |
| Fifteen negative controls | 1,500/1,500 rejected |
No older or partial model score is carried onto v3.2. A leaderboard row is published only after one model runs all 100 tasks on this exact release.
MCP fidelity and licenses
The world maps each allowlisted tool to a documented Clio Manage v4, Gmail v1, Google Drive v3, or Slack Web API operation. Contract metadata and source links ship with the release. Synthetic data is CC-BY-4.0; benchmark code is Apache-2.0. Every entity, person, event, amount, and address is fictitious.