SkillsBench is a benchmark for measuring how effectively AI agents use modular skills—folders containing instructions, scripts, and resources—to complete specialized tasks. It helps researchers and developers evaluate both skill quality and agent behavior, including tasks that require combining multiple skills. The catalogue’s skills and instructions are evaluated as part of this workflow.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add benchflow-ai/skillsbench --skill gpt-multimodalgit clone --depth 1 https://github.com/benchflow-ai/skillsbenchWrote 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.
[](https://agentmods.dev/skills/benchflow-ai/skillsbench/gpt-multimodal)<a href="https://agentmods.dev/skills/benchflow-ai/skillsbench/gpt-multimodal"><img src="https://agentmods.dev/badge/skills/benchflow-ai/skillsbench/gpt-multimodal/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/benchflow-ai/skillsbench/gpt-multimodal"><img src="https://agentmods.dev/badge/skills/benchflow-ai/skillsbench/gpt-multimodal.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00018 | $0.04425 |
| Opus 5 | $0.00009 | $0.02212 |
| Sonnet 5 | $0.00004 | $0.00885 |
| Haiku 4.5 | $0.00002 | $0.00443 |
Grade A, and why
gpt-multimodal scanned grade A with 0 findings 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 10d ago.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
Copies of this mod
1 near-identical copy found in the catalogue:
- gpt-multimodal — 100% identical, 41 lines differ
How it starts
The opening of the file, as written. The whole thing — 649 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OpenAI Vision Analysis Skill
Purpose
This skill enables image analysis, scene understanding, text extraction, and multi-frame comparison using OpenAI's vision-capable GPT models (e.g., gpt-4o, gpt-5). It supports single and multiple images analysis and sequential frames for temporal analysis.
When to Use
- Analyzing image content (objects, scenes, colors, spatial relationships)
- Extracting and reading text from images (OCR via vision models)
- Comparing multiple images to detect differences or changes
- Processing video frames to understand temporal progression
- Generating detailed image descriptions or captions
- Answering questions about visual content
Required Libraries
The following Python libraries are required:
from openai import OpenAI
import base64
import json
import os
from pathlib import Path
Input Requirements
- File formats: JPG, JPEG, PNG, WEBP, non-animated GIF
- Image quality: Clear and legible; minimum 512×512px recommended
- File size: Under 20MB per image recommended
- Maximum per request: Up to 500 images, 50MB total payload
- URL or Base64: Images can be provided as URLs or base64-encoded data
Output Schema
All analysis results should be returned as valid JSON conforming to this schema:
{
"success": true,
"model": "gpt-5",
"analysis": "Detailed description or analysis of the image content...",
"metadata": {
"image_count": 1,
"detail_level": "high",
"tokens_used": 850,
"processing_time_ms": 1234
},
"extracted_data": {
"objects": ["car", "person", "building"],
"text_found": "Sample text from image",
"colors": ["blue", "white", "gray"],
"scene_type": "urban street"
},
"warnings": []
}
Field Descriptions
success: Boolean indicating whether the API call succeededmodel: The GPT model used for analysis (e.g., "gpt-4o", "gpt-5")analysis: Complete textual analysis or description from the modelmetadata.image_count: Number of images analyzed in this requestmetadata.detail_level: Detail parameter used ("low", "high", or "auto")metadata.tokens_used: Approximate token count for the requestmetadata.processing_time_ms: Time taken to process the requestextracted_data: Structured information extracted from the image(s)warnings: Array of issues or limitations encountered
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.
- 10d ago First seen · 649 lines · 18 tokens per session scan A faf77d38e4ed
gpt-multimodal is a skill published in the GitHub repository benchflow-ai/skillsbench (1,757 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 18 tokens to every session and 4,425 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
rag-eval
NVIDIA RAG Blueprint evaluation guidance for measuring retrieval and answer quality with stable datasets, baselines, and reproducible scoring workflows.
local-llm-ops
Local LLM operations with Ollama on Apple Silicon, including setup, model pulls, chat launchers, benchmarks, and diagnostics.
fixed-tensor-testing
Test ML functions with fixed input tensors for reproducibility.
dbscan-custom-metric
Run DBSCAN clustering with a custom distance metric using sklearn, including how to define weighted Euclidean metrics and extract cluster centroids.
parallel-grid-search
Parallelize hyperparameter grid search using joblib for efficient multi-core execution.
json-data-extraction
Extract, parse, and query JSON data from large enterprise files efficiently.