mediause/agent-skills/huggingface/SKILL.md
huggingface
Standardized Hugging Face skill for MediaUse. Guest-only mode with no login requirement, strict context flow, full dynamic command map, and safety/rate controls.
- Source repository stars
- 9
- Declared platforms
- 0
- Static risk flags
- 2
- Last source update
- 2026-07-28
- Source checked
- 2026-07-28
Decision brief
What it does—and where it fits
Use Hugging Face discovery workflows in MediaUse with a guest-only, no-login flow for papers, models, datasets, and spaces.
Not for
- Tasks that require unconfirmed production actions or broad system permissions.
- Environments where the pinned source and install steps cannot be inspected.
Compatibility matrix
Platform support, with evidence labels
| Platform | Status | Evidence | What to check |
|---|---|---|---|
| Codex | Not declared | No explicit evidence | Portability before use |
| Claude Code | Not declared | No explicit evidence | Portability before use |
| Cursor | Not declared | No explicit evidence | Portability before use |
| Gemini CLI | Not declared | No explicit evidence | Portability before use |
Installation
Inspect first. Install second.
The source command is displayed only when detected. A safe inspection prompt is always available so your agent can explain every action before execution.
npx skills add https://github.com/mediause/agent-skills --skill "huggingface"Inspect the Agent Skill "huggingface" from https://github.com/mediause/agent-skills/blob/766e5075b28b3acd541436a7e3b3e21cdc38da01/huggingface/SKILL.md at commit 766e5075b28b3acd541436a7e3b3e21cdc38da01. List every install step, command, network request, credential, file read/write, external action, and rollback step. Explain whether it fits my task. Do not install or execute anything until I approve.
Workflow
What the source asks the agent to do
- 01
3.1 Discover and plugin setup
Review the “3.1 Discover and plugin setup” section in the pinned source before continuing.
Review and apply the “3.1 Discover and plugin setup” source section. - 02
5. Workflow Examples
Review the “5. Workflow Examples” section in the pinned source before continuing.
Review and apply the “5. Workflow Examples” source section. - 03
always run once before each workflow (auto-upgrade latest)
powershell -C "iwr https://release.mediause.dev/install.ps1 -UseBasicParsing | iex" mediause --version
powershell -C "iwr https://release.mediause.dev/install.ps1 -UseBasicParsing | iex" mediause --version - 04
Scope
Use this skill when the task targets Hugging Face operations such as:
Account: plugin health checkPapers: top paper rankings, single paper detailModels: ranked model listing - 05
1. Install MediaUse CLI (Windows Only)
Use the official install script for Windows:
https://release.mediause.dev/install.ps1Before every run, execute the install script once to auto-install or auto-upgrade to the latest MediaUse CLI.Windows: supported
Permission review
Static risk signals and limitations
Runs scripts
The documentation asks the agent to run terminal commands or scripts.
Before every run, execute the install script once to auto-install or auto-upgrade to the latest MediaUse CLI.Network access
The documentation includes network, browsing, or remote request actions.
powershell -C "iwr https://release.mediause.dev/install.ps1 -UseBasicParsing | iex"Runs scripts
The documentation asks the agent to run terminal commands or scripts.
First step on every run: execute install script once to auto-install/auto-upgrade the latest MediaUse CLI.Network access
The documentation includes network, browsing, or remote request actions.
powershell -C "iwr https://release.mediause.dev/install.ps1 -UseBasicParsing | iex"Evidence record
Why each signal appears
| Signal | Value | Evidence type | Meaning |
|---|---|---|---|
| Quality score | 75/100 | Computed | Documentation, specificity, maintenance, and trust rules |
| Repository stars | 9 | Source | Repository attention, not individual Skill quality |
| Compatibility | 0 platforms | Source | Declared in the catalog source record |
| Usage guide | automated source guide | Editorial | Generated or reviewed according to the visible evidence level |
Pinned source
Provenance and original SKILL.md
- Repository
- mediause/agent-skills
- Skill path
- huggingface/SKILL.md
- Commit
- 766e5075b28b3acd541436a7e3b3e21cdc38da01
- License
- Not declared
- Collected
- 2026-07-28
- Default branch
- master
View the original SKILL.md
MediaUse Hugging Face Skill
Use Hugging Face discovery workflows in MediaUse with a guest-only, no-login flow for papers, models, datasets, and spaces.
Scope
Use this skill when the task targets Hugging Face operations such as:
- Account: plugin health check
- Papers: top paper rankings, single paper detail
- Models: ranked model listing
- Datasets: ranked dataset listing
- Spaces: ranked space listing
Hugging Face in this plugin is guest-only. No login is required.
1. Install MediaUse CLI (Windows Only)
Use the official install script for Windows:
-
https://release.mediause.dev/install.ps1 Mandatory update rule:
-
Before every run, execute the install script once to auto-install or auto-upgrade to the latest MediaUse CLI.
Run:
powershell -C "iwr https://release.mediause.dev/install.ps1 -UseBasicParsing | iex"
Then verify:
mediause --version
Current support status:
- Windows: supported
- Linux: not supported yet
- macOS: not supported yet
Recommended skill install path:
- .mediause/skills/huggingface/SKILL.md
2. Get and Configure MediaUse Key
2.1 Apply for key
- Open https://mediause.dev/
- Sign in to your account.
- Open Project.
- Create or copy your API key.
2.2 Configure key in CLI
mediause manage key <your_key> --json
3. Core Flow (Guest-Only, No Login)
Always follow this order:
First step on every run: execute install script once to auto-install/auto-upgrade the latest MediaUse CLI.
- Discover site and commands.
- Bind guest context with
use account huggingface:guest. - Execute dynamic read actions.
- Verify with trace/task.
Important constraints:
- Hugging Face plugin here only uses guest mode.
- Do not run
mediause auth login huggingface. - Do not require
mediause auth healthfor this plugin flow.
3.1 Discover and plugin setup
mediause plugin list --json
mediause plugin add huggingface --json
mediause huggingface -h
mediause huggingface get -h
mediause huggingface account -h
3.2 Bind guest context
mediause use account huggingface:guest --json
If the page/session is challenged (for example unusual traffic), rerun with visible mode:
mediause use account huggingface:guest --show --json
Risk reduction note:
--showusually lowers repeated anti-bot interception risk in challenged sessions.
4. Hugging Face Dynamic Command Map (v1)
Source manifest:
- plugin:
plugin.huggingface - file:
crates/platforms/plugins/huggingface/manifest.yaml
4.1 account.*
mediause huggingface account health --json
4.2 get.top
Top upvoted Hugging Face papers.
mediause huggingface get top [--limit <n>] [--all <bool>] [--date <YYYY-MM-DD>] [--period <daily|weekly|monthly>] --json
Notes:
--perioddefault isdaily--dateis for daily mode--all trueignores limit and returns full ranked set
4.3 get.models
Top Hugging Face models.
mediause huggingface get models [--sort <downloads|likes|trending|created_at|last_modified>] [--search <text>] [--pipeline <tag>] [--limit <n>] --json
4.4 get.datasets
Top Hugging Face datasets.
mediause huggingface get datasets [--sort <downloads|likes|trending|created_at|last_modified>] [--search <text>] [--limit <n>] --json
4.5 get.spaces
Top Hugging Face spaces.
mediause huggingface get spaces [--sort <likes|created_at|last_modified>] [--search <text>] [--sdk <gradio|streamlit|docker|static>] [--limit <n>] --json
4.6 get.paper
Hugging Face paper detail by arXiv id.
mediause huggingface get paper --id <arxiv_id> --json
Example ids:
1706.037622401.06196
5. Workflow Examples
5.1 Daily paper scan
mediause plugin add huggingface --json
mediause use account huggingface:guest --json
mediause huggingface get top --period daily --limit 20 --json
mediause trace last --json
5.2 Weekly leaderboard and model follow-up
mediause use account huggingface:guest --json
mediause huggingface get top --period weekly --limit 50 --json
mediause huggingface get models --sort downloads --search llama --limit 20 --json
mediause trace last --json
5.3 Dataset and space discovery
mediause use account huggingface:guest --json
mediause huggingface get datasets --sort likes --search multilingual --limit 20 --json
mediause huggingface get spaces --sort likes --sdk gradio --limit 20 --json
mediause trace last --json
5.4 Paper deep read
mediause use account huggingface:guest --json
mediause huggingface get paper --id 1706.03762 --json
mediause trace last --json
6. Operational Constraints (Mandatory)
6.1 Read-only and guest-only
- This plugin flow is read-only.
- Use
huggingface:guestonly. - Do not attempt write actions or login flow for this plugin.
6.2 Frequency limits and spacing
Suggested limits:
get top/models/datasets/spaces: <= 60 calls per minuteget paper: <= 30 calls per minute
Minimum spacing:
- list/ranking calls: >= 1 second between calls
- single paper detail calls: >= 2 seconds between calls
If throttled or blocked:
- Pause 30-60 seconds.
- Retry with lower call frequency.
- Re-bind guest context with
--showif challenge is present.
6.3 Safety policy
- Do not bypass anti-bot or platform protections.
- Do not scrape credentials or session secrets.
- Respect Hugging Face usage policies and local regulations.
6.4 Output and error handling
- Prefer
--jsonoutput for automation. - Keep stable machine-readable handling for errors.
- On HTTP 429 or challenge pages, stop and retry with cooldown.
7. Execution Checklist
Before run:
Before every run, execute the install script once.
- CLI installed via
https://release.mediause.dev/install.ps1on Windows. - PATH updated and
mediause --versionworks. - API key configured.
- Plugin installed and help output verified.
- Guest context bound using
mediause use account huggingface:guest --json.
During run:
- Use only read commands listed in this skill.
- Keep request spacing and avoid burst loops.
- Use
--showonly when challenge handling is needed.
After run:
- Save outputs/logs.
- Record retry/cooldown events.
- Keep traffic conservative for repeated polling jobs.
8. Quick Command Reference
# always run once before each workflow (auto-upgrade latest)
powershell -C "iwr https://release.mediause.dev/install.ps1 -UseBasicParsing | iex"
mediause --version
# discover
mediause plugin list --json
mediause plugin add huggingface --json
mediause huggingface -h
mediause huggingface get -h
# guest context
mediause use account huggingface:guest --json
# actions
mediause huggingface account health --json
mediause huggingface get top --period daily --limit 20 --json
mediause huggingface get models --sort downloads --limit 20 --json
mediause huggingface get datasets --sort likes --limit 20 --json
mediause huggingface get spaces --sort likes --limit 20 --json
mediause huggingface get paper --id 1706.03762 --json
# trace
mediause trace last --json
Skill Metadata Maintainer: @mediause-demo Last-Updated: 2026-05-21 Version: v1
Alternatives
Compare before choosing
event4u-app/agent-config
design-review
Use when the user says "review the design", "check the UI", or wants a comprehensive UI/UX review. Uses a 7-phase methodology covering interaction, responsiveness, accessibility, and more.
K-Dense-AI/scientific-agent-skills
dask
Distributed computing for larger-than-RAM pandas/NumPy workflows. Use when you need to scale existing pandas/NumPy code beyond memory or across clusters. Best for parallel file processing, distributed ML, integration with existing pandas code. For out-of-core analytics on single machine use vaex; for in-memory speed use polars.
K-Dense-AI/scientific-agent-skills
medchem
Medicinal chemistry filters for compound triage. Apply drug-likeness rules (Lipinski, Veber, CNS), structural alert catalogs (PAINS, NIBR, ChEMBL), complexity metrics, and the medchem query language for library filtering.
K-Dense-AI/scientific-agent-skills
neurokit2
Use NeuroKit2 to build or audit reproducible research workflows for physiological time-series preprocessing, event/interval analysis, multimodal alignment, variability, and complexity. Trigger when code imports neurokit2 or needs its current APIs, schemas, and method-aware validation—not for diagnosis or device validation.