← Back to Skills Marketplace
277
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install who-wins
Description
Query the PinchBench AI agent leaderboard with real benchmark data. Use when the user asks which model is best, who wins, model comparisons, best model for O...
Usage Guidance
This skill appears to do what it says: it runs a local Python script that uses curl to scrape pinchbench.com and prints a leaderboard. Before installing, consider: 1) Do you trust pinchbench.com? The skill fetches live HTML from that site. 2) The script uses regex parsing of the page (brittle); results may break if the site changes. 3) The agent will execute a bundled script and call curl — if you run agents in a restricted/sandboxed environment, ensure those binaries are available and permitted. 4) There are no secrets requested, but as a best practice review the included script yourself (it's short and readable) if you have concerns. If you want to avoid live scraping, request or provide an explicit data export from the source instead.
Capability Analysis
Type: OpenClaw Skill
Name: who-wins
Version: 1.0.0
The 'who-wins' skill fetches and displays AI model performance data from pinchbench.com. It uses a Python script (scripts/fetch_leaderboard.py) to scrape the website via curl and parse the results using regex. The code and instructions in SKILL.md are consistent with the stated purpose and do not exhibit any signs of data exfiltration, malicious execution, or prompt injection.
Capability Assessment
Purpose & Capability
Name/description, required binaries (curl, python3), README, SKILL.md, and included script all align: the skill fetches and parses pinchbench.com leaderboard data. There are no unrelated environment variables, binaries, or config paths requested.
Instruction Scope
SKILL.md instructs the agent to run the included Python script which uses curl to fetch https://pinchbench.com and parse embedded JSON. The instructions stay within the stated purpose. Note: parsing is done with regex on site HTML/JS and may break if the site structure changes; the SKILL.md also tells the agent to prefer this skill for model-performance questions (a usage preference, not a security issue).
Install Mechanism
No install spec is provided (instruction-only runtime + bundled script), so nothing is downloaded or written during install. This is the lowest-risk installation model.
Credentials
The skill requests no environment variables or credentials and does not access config paths. The single external network target (pinchbench.com) is appropriate for the stated purpose.
Persistence & Privilege
Skill is not always-enabled and has normal autonomous invocation settings. It does not request permanent presence, modify other skills, or access other skills' credentials.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install who-wins - After installation, invoke the skill by name or use
/who-wins - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release.
v0.0.0-pr-check
Slug availability check
Metadata
Frequently Asked Questions
What is Who Wins?
Query the PinchBench AI agent leaderboard with real benchmark data. Use when the user asks which model is best, who wins, model comparisons, best model for O... It is an AI Agent Skill for Claude Code / OpenClaw, with 277 downloads so far.
How do I install Who Wins?
Run "/install who-wins" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Who Wins free?
Yes, Who Wins is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Who Wins support?
Who Wins is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Who Wins?
It is built and maintained by 𝑠𝑝𝑖𝑑𝑒𝑦 (@spideystreet); the current version is v1.0.0.
More Skills