CtrlK
BlogDocsLog inGet started
Tessl Logo

huggingface-lora-space-builder

Build and publish a Gradio demo on Hugging Face Spaces for a user-provided LoRA. Use when someone asks to create, generate, ship, or publish a Space, demo, Gradio app, or playground for a LoRA — including LoRAs for Qwen-Image, Qwen-Image-Edit, LTX-Video, Wan, FLUX, SDXL, or other diffusion base models. Also triggers when someone describes a LoRA they trained or hosts on the Hub and wants to share it. Covers picking the right base pipeline and `diffusers` inference recipe, designing a UI tailored to the LoRA's task and inputs (Union/multi-task control, edit, video, image, etc.), respecting model-card recommendations (trigger words, steps, guidance, LoRA scale, example inputs), and shipping to ZeroGPU hardware as a private Space by default.

75

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

High

Do not use without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A strong, highly actionable operating manual: the phased workflow has real validation checkpoints and error-recovery loops, and nearly all guidance is concrete and executable. The main costs are repetition of the batching/approval discipline and a body length that carries detail better suited to the (excellent) reference files.

Suggestions

State the 'batch questions / batched single approval' rule once, in the Workflow section, and delete its restatements in Phase 1, the Phase 4 opening, the 'Single batched approval' subsection, and 'What to avoid'.

Move the xformers FA3 Blackwell workaround and the gr.Examples fn TypeError diagnosis into references/zerogpu-and-publishing.md (or a pitfalls reference), leaving a one-line pointer plus the rule in the body.

Trim the 'What "good" looks like' section to its concrete bullet list, cutting the philosophical framing ('The demo should feel handcrafted...') that the bullets already express.

DimensionReasoningScore

Conciseness

The body is dense with skill-specific, non-obvious knowledge (ZeroGPU rules, pipeline-class pitfalls, cache_mode="lazy", the xformers FA3 Blackwell bug), but the batched-approval discipline and question-batching rules are each restated three or more times across the intro, phase openings, and 'What to avoid'. Mostly efficient with trimmable repetition rather than padding — fits anchor 4, not 5.

4 / 5

Actionability

Fully executable throughout: working code for the token check, create_repo/upload, and build-polling loop; concrete `gradio predict` JSON payloads; the README YAML template; and error-message-to-fix mappings for publish-time, build-time, and inference failures. Copy-paste ready and covers the common cases.

5 / 5

Workflow Clarity

Six phases are explicitly sequenced with a summary workflow list, a mandatory Phase 2 pipeline-class verification checkpoint, batched user-approval gates, and a Phase 6 smoke-test with explicit feedback loops (503 → wake and retry; weight_name error → re-check list_repo_files; ImportError → add dep and re-upload).

5 / 5

Progressive Disclosure

References are well-signaled and one level deep, and every referenced path (tasks.md, adapting-to-the-lora.md, creative-mode.md, zerogpu-and-publishing.md, base-models/*.md) exists in the bundle. However, the ~394-line body inlines detail that could live in a reference — the xformers FA3 dispatch workaround and the gr.Examples fn TypeError fix — keeping it just short of the clean split of anchor 5.

4 / 5

Total

18

/

20

Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description explicitly answers both what it does and when to use it, with multiple concrete capabilities, comprehensive natural trigger terms including model names, and a distinct niche. Its only weakness is that it runs long for a description field, though every clause carries triggering or scoping information.

DimensionReasoningScore

Specificity

'Build and publish a Gradio demo on Hugging Face Spaces', 'picking the right base pipeline and `diffusers` inference recipe', 'designing a UI tailored to the LoRA's task and inputs', 'respecting model-card recommendations (trigger words, steps, guidance, LoRA scale, example inputs)', and 'shipping to ZeroGPU hardware as a private Space' — multiple specific concrete actions with comprehensive coverage, matching anchor 5.

5 / 5

Completeness

Both questions are answered explicitly and concretely: the 'what' in the opening sentence and the 'Covers...' clause, and the 'when' in an explicit 'Use when...' clause plus a second 'Also triggers when...' trigger case.

5 / 5

Trigger Term Quality

Natural user phrasing is covered comprehensively with synonyms: 'create, generate, ship, or publish a Space, demo, Gradio app, or playground for a LoRA', base-model names users actually say (Qwen-Image, Qwen-Image-Edit, LTX-Video, Wan, FLUX, SDXL), and 'describes a LoRA they trained or hosts on the Hub'.

5 / 5

Distinctiveness Conflict Risk

A clear niche — LoRA demo Spaces on Hugging Face hardware — with distinct trigger phrases (LoRA + Space/demo/Gradio/playground + named diffusion base models); it is unlikely to fire for unrelated Gradio, Hub, or general deployment work.

5 / 5

Total

20

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 3 deeper-than-1-level

Warning

Total

15

/

16

Passed

Repository
huggingface/skills
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.