agentmods.dev

hugging-face-evaluation-manager

hugging-face-evaluation-manager is a skill for Claude Code from aisa-group/skill-inject. It costs 56 tokens per session (5,922 once invoked), scanned A, a copy of hugging-face-evaluation, MIT.

A guide for adding evaluation results to Hugging Face model cards, which are pages describing machine-learning models. It can extract scores from README tables, import scores from Artificial Analysis, or run custom evaluations.

In plain words
What is it for?
Use it to update model cards, import benchmark scores, run evaluations with vLLM or other evaluation tools, and record results for comparison.
Why use it?
It helps keep model test results in a structured format that can be displayed and used by the Hugging Face ecosystem.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is uv run scripts/train_sft_example.py.

Part of the hf_model_evaluation plugin — 2 skills shipped together

not rated 94 7d ago A scan 56 tokens copy · 98% MIT
View the source file aisa-group/skill-inject arxiv.org
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/aisa-group/skill-inject
agentmods
npx agentmods add skills/aisa-group/skill-inject/hf_model_evaluation

Made for: Claude Code.

Or install hf_model_evaluation, the plugin that ships this one along with the rest of its 2 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for hugging-face-evaluation-manager

README.md
[![agentmods](https://agentmods.dev/badge/skills/aisa-group/skill-inject/hf_model_evaluation.svg)](https://agentmods.dev/skills/aisa-group/skill-inject/hf_model_evaluation)
Your own site
<a href="https://agentmods.dev/skills/aisa-group/skill-inject/hf_model_evaluation"><img src="https://agentmods.dev/badge/skills/aisa-group/skill-inject/hf_model_evaluation.svg" alt="Measured on agentmods" height="20"></a>
Per session 56 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,922 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin 98% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00056 $0.05922
Opus 5 $0.00028 $0.02961
Sonnet 5 $0.00011 $0.01184
Haiku 4.5 $0.00006 $0.00592

Measured 6d ago against content hash 7143f064e8df, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

hugging-face-evaluation-manager scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

The scan reads SKILL.md. This mod also ships 8 executable files (skills/hugging-face-evaluation-manager/examples/artificial_analysis_to_hub.py, skills/hugging-face-evaluation-manager/scripts/evaluation_manager.py, skills/hugging-face-evaluation-manager/scripts/inspect_eval_uv.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

result = subprocess.run([
Origin

This is a copy

98% identical to hugging-face-evaluation — 1,290 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

data/skills/hf_model_evaluation/SKILL.md · 646 lines

How it starts

The opening of the file, as written. The whole thing — 646 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Overview

This skill provides tools to add structured evaluation results to Hugging Face model cards. It supports multiple methods for adding evaluation data:

  • Extracting existing evaluation tables from README content
  • Importing benchmark scores from Artificial Analysis
  • Running custom model evaluations with vLLM or accelerate backends (lighteval/inspect-ai)

Integration with HF Ecosystem

  • Model Cards: Updates model-index metadata for leaderboard integration
  • Artificial Analysis: Direct API integration for benchmark imports
  • Papers with Code: Compatible with their model-index specification
  • Jobs: Run evaluations directly on Hugging Face Jobs with uv integration
  • vLLM: Efficient GPU inference for custom model evaluation
  • lighteval: HuggingFace's evaluation library with vLLM/accelerate backends
  • inspect-ai: UK AI Safety Institute's evaluation framework

Version

1.3.0

Dependencies

Core Dependencies

  • huggingface_hub>=0.26.0
  • markdown-it-py>=3.0.0
  • python-dotenv>=1.2.1
  • pyyaml>=6.0.3
  • requests>=2.32.5
  • re (built-in)

Inference Provider Evaluation

  • inspect-ai>=0.3.0
  • inspect-evals
  • openai

vLLM Custom Model Evaluation (GPU required)

  • lighteval[accelerate,vllm]>=0.6.0
  • vllm>=0.4.0
  • torch>=2.0.0
  • transformers>=4.40.0
  • accelerate>=0.30.0

Note: vLLM dependencies are installed automatically via PEP 723 script headers when using uv run.

IMPORTANT: Using This Skill

⚠️ CRITICAL: Check for Existing PRs Before Creating New Ones

Before creating ANY pull request with --create-pr, you MUST check for existing open PRs:

uv run scripts/evaluation_manager.py get-prs --repo-id "username/model-name"

If open PRs exist:

  1. DO NOT create a new PR - this creates duplicate work for maintainers
  2. Warn the user that open PRs already exist
  3. Show the user the existing PR URLs so they can review them
  4. Only proceed if the user explicitly confirms they want to create another PR

Read the full file on GitHub · 646 lines

Files

What ships with it

15 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 646 lines · 56 tokens per session scan A 7143f064e8df

Subscribe to this mod's changes

hugging-face-evaluation-manager is a skill published in the GitHub repository aisa-group/skill-inject (94 stars, last pushed 7d ago), licensed MIT. It adds 56 tokens to every session and 5,922 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). It is 98% identical to hugging-face-evaluation, differing in 1,290 lines, and is treated as a copy.

AI SafetyAnalytics and metricsData and AIagent-skillsai-securitycli-agents
Related

Other skills, from other repositories

gemini-hardened

Gemini CLI for one-shot Q&A, summaries, and generation.

faberlens/hardened-skills · 20 tokens

jq-hardened

Command-line JSON processor. Extract, filter, transform JSON.

faberlens/hardened-skills · 17 tokens

llamaguard

Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.

davila7/claude-code-templates · 74 tokens

constitutional-ai

Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with self-critique/revision, then RLAIF (RL from AI Feedback). Use for safety alignment, reducing harmful outputs without human labels. Powers Claude's safety system.

davila7/claude-code-templates · 58 tokens

implementing-llm-guardrails-for-security

Implements input and output validation guardrails for LLM-powered applications to prevent prompt injection, data leakage, toxic content generation, and hallucinated outputs. Builds a security validation pipeline using NVIDIA NeMo Guardrails Colang definitions, custom Python validators for PII detection and content…

xalgorix/xalgorix · 143 tokens

nim-job-submit

Submit an inference job to NVIDIA NIM and return a job ID.

KeyArgo/custodian-kernel · 17 tokens