# NVIDIA/SkillEvaluator

> Open-source NVIDIA tool for evaluating AI agent skills defined by the Agent Skills specification. It runs static quality gates, security scanning, overlap detection and sandboxed live agent runs to measure how a skill affects task results.

- Magnitude: 5.9 out of 10 — Early signal
- Stars: 1,171 total · +49 stars measured 2026-11-14, 00:03–11:13 UTC, ≈ 131 by evening
- Star trust: star growth looks organic
- Category: AI agents · Language: Python · License: Apache-2.0 · Created: 2026-06-24 · Last push: 2026-11-14
- GitHub: https://github.com/NVIDIA/SkillEvaluator · Homepage: https://docs.nvidia.com/skills/skillevaluator/ · Page: https://gitnova.dev/en/r/NVIDIA/SkillEvaluator

## Useful for

- Check changed agent skills in a GitHub or GitLab pipeline before merge
- Find duplicated or semantically overlapping instructions between skills
- Compare sandboxed agent runs with and without a skill to measure its impact

## How to get started

```sh
git clone --depth 1 https://github.com/NVIDIA/SkillEvaluator.git
cd SkillEvaluator
```

Copied as is from the “quickstart” section of the README. GitNova has not run these commands: check them in the original before running: https://github.com/NVIDIA/SkillEvaluator#readme

## Why it’s here

- Star-counter measurements on 2026-11-14 (UTC), 00:03–11:13: 1122 → 1171 stars (+49). This is the change over that interval.
- Estimated end-of-day forecast: about +131 stars, using observed gains and the previous day.
- Over the last two days the pace is 14× that of the previous week and a half.
- Elevated gains were measured on the last 3 completed UTC days in a row. Continued growth is not guaranteed.
- About 24 forks a day — people are taking the code.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 135
- Issues and pull requests: 547
- Watchers: 1
- Average over the last week: 90 per day
- Usual pace: 4 per day
- Stars in the last hour (measured): 5
- Latest release: v0.9.0 (2026-11-13)

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-10-16 … 2026-11-14: 1, 1, 3, 5, 5, 2, 5, 1, 2, 3, 3, 7, 4, 1, 3, 1, 5, 5, 5, 8, 7, 5, 5, 5, 9, 3, 71, 207, 203, 49

## Magnitude by day, last 3 days

11-12 4.9, 11-13 6.4, 11-14 6.0

## More in this category

1. **deepseek-ai/chat-demo** — 8.4 · Early signal · AI agents · TypeScript · +709 stars measured 2026-11-14, 00:03–11:13 UTC, ≈ 1,416 by evening
   A static chat demo powered by the DeepSeek API that runs entirely in the browser without a backend; users enter their own API key to chat with streaming replies, attachments, and speech features.
   Full card: https://gitnova.dev/en/r/deepseek-ai/chat-demo.md
2. **GoogleChrome/modern-web-guidance-src** — 8.1 · Breakout · AI agents · TypeScript · +474 stars measured 2026-11-14, 00:03–11:13 UTC, ≈ 1,113 by evening
   A set of skills and a CLI that embed modern web platform APIs, best practices, and browser compatibility data into coding agents so they stop generating legacy code.
   Full card: https://gitnova.dev/en/r/GoogleChrome/modern-web-guidance-src.md
3. **ZhuLinsen/daily_stock_analysis** — 7.6 · Breakout · AI agents · Python · +465 stars measured 2026-11-14, 00:03–11:13 UTC, ≈ 925 by evening
   LLM-powered multi-market stock analysis system: gathers quotes, news and fundamentals for A-shares, Hong Kong, US, Japan, Korea and Taiwan, produces a daily decision dashboard and pushes it to messengers. Runs on a schedule in GitHub…
   Full card: https://gitnova.dev/en/r/ZhuLinsen/daily_stock_analysis.md
4. **ChromeDevTools/chrome-devtools-mcp** — 7.5 · Breakout · AI agents · TypeScript · +414 stars measured 2026-11-14, 00:04–11:13 UTC, ≈ 781 by evening
   An MCP server from Chrome DevTools that lets coding agents control a live Chrome browser: record performance traces, inspect network requests, take screenshots and read console messages. A CLI is also provided for use without MCP.
   Full card: https://gitnova.dev/en/r/ChromeDevTools/chrome-devtools-mcp.md
5. **backnotprop/plannotator** — 7.0 · Breakout · AI agents · TypeScript · +201 stars measured 2026-11-14, 00:03–11:13 UTC, ≈ 407 by evening
   A local browser-based tool for visually annotating plans, documents, and code diffs from AI coding agents (Claude Code, Codex, Copilot CLI, etc.) and sending feedback back to the agent.
   Full card: https://gitnova.dev/en/r/backnotprop/plannotator.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-11-14 11:22 UTC, updated every 30 minutes.