LaunchKit · 2026
Back to Skills

pinchbench

Run PinchBench benchmarks to evaluate OpenClaw agent performance across real-world tasks.

0
439 downloads
by @olearycrew

Setup & Installation

openclaw skills install @olearycrew/pinchbench

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install pinchbench

What This Skill Does

PinchBench measures LLM model performance as the brain of an OpenClaw agent across 23 standardized tasks. Tasks span calendar, email, research, coding, file management, and multi-step workflows. Results are graded automatically and can be submitted to a public leaderboard at pinchbench.com.

Provides 23 reproducible, auto-graded tasks with a public leaderboard, making model comparisons objective rather than anecdotal.

When to use it

  • Comparing two models on email triage and calendar tasks before picking one
  • Running the automated-only suite for a quick performance check before a release
  • Submitting benchmark scores to the PinchBench public leaderboard
  • Testing a specific subset of tasks like market research or PDF summarization
  • Verifying a new OpenClaw setup handles multi-step API workflows correctly

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: How well does claude-sonnet-4 handle email and calendar tasks?

AGENT
  1. 1Register for a PinchBench API token using the --register flag
  2. 2Run the benchmark targeting specific tasks: --suite task_01_calendar,task_16_email_triage,task_17_email_search
  3. 3Wait for tasks to complete and results to save as JSON in the results/ directory
  4. 4Inspect scores with jq to extract per-task grading means
  5. 5Results upload automatically to the leaderboard using the registered token
OUTPUT

Per-task scores for calendar and email tasks, plus an overall average, visible on the pinchbench.com leaderboard

Requirements

Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.

PinchBench API token for leaderboard submission (register once via --register flag)