CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

comfyui

Use when working with ComfyUI workflows in OpenMontage, including comfyui_image/comfyui_video/comfyui_music, custom workflow_json/workflow_path inputs, output_node selection, missing model setup, LoRAs, low-VRAM workflow choices, and community workflow imports.

67

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

90%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.

An exemplary instruction-only skill body: lean, non-redundant, and packed with tool-contract specifics Claude could not infer, with concrete error-recovery loops (resume_prompt_id, missing_models) and a genuinely valuable trap warning about video latent nodes. The only gaps are the absence of a consolidated ordered pre-flight checklist and slight organizational roughness in section placement.

DimensionReasoningScore

Conciseness

The body is dense with non-inferable contract details Claude could not know — the resume_prompt_id timeout recovery, the EmptyLatentImage batch_size strobing trap, Partner Node hosting requirements, ACE-Step v1 native-node specifics — with zero concept explanations or padding. Every sentence adds operational information; it matches anchor 5's 'every token earns its place'.

5 / 5

Actionability

Guidance is fully concrete and executable: exact env vars (COMFYUI_IMAGE_SERVER_URL, COMFYUI_MUSIC_SERVER_URL), endpoints (GET /system_stats, POST /prompt, GET /history/{prompt_id}), specific node class names (EmptyHunyuanLatentVideo, Wan22ImageToVideoLatent, SaveAudioMP3), a literal output_node example ('"108"'), folder paths, and structured payload fields (data.missing_models[]). It covers the common cases for all three tools plus failure paths.

5 / 5

Workflow Clarity

Explicit checkpoints exist ('Check the latent node before running any unfamiliar video workflow', check ComfyUI history before retrying from scratch) and a real error-recovery feedback loop (resume_prompt_id, missing_models install flow). However, the overall pre-flight sequence is scattered across topical sections rather than presented as an ordered workflow or consolidated checklist, which keeps it below anchor 5.

4 / 5

Progressive Disclosure

Well-organized ##/### sections with no nested references and no content that clearly belongs in a separate file at this length (~78 lines). It falls short of anchor 5 because there is no reference structure at all — the skill is over the 50-line simple-skill threshold and has minor placement awkwardness (Partner Nodes nested under Server Contract, Music appended after Failure Handling) — but navigation is easy and nothing is buried.

4 / 5

Total

18

/

20

Passed

Description

70%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.

A highly distinctive, trigger-rich description that names the exact tools and subtopics (output_node, LoRAs, VRAM, community imports), but it is structured as a single 'Use when...' clause with no explicit 'what' statement — the skill's actual actions are only implied by the topic list. Trigger terms are natural for the domain but miss common generation phrasings and model family names.

Suggestions

Add an explicit 'what' statement with action verbs before the 'Use when' clause, e.g. 'Guides ComfyUI generation through OpenMontage's comfyui_image/comfyui_video/comfyui_music tools: workflow selection, custom workflow_json/workflow_path inputs, output_node choice, and model/LoRA setup. Use when...' — this would lift completeness from 3 to 5.

Include natural trigger phrases users would actually say, such as 'generate an image, video, or song with ComfyUI' or 'import a community ComfyUI workflow', to improve trigger_term_quality.

Add one or two model family names users commonly mention (e.g. Wan, LTXV, ACE-Step) as trigger synonyms, since these are the terms a user requesting generation is most likely to use.

DimensionReasoningScore

Specificity

The description lists several specific concrete items — 'custom workflow_json/workflow_path inputs, output_node selection, missing model setup, LoRAs, low-VRAM workflow choices, and community workflow imports' — but phrases them as topics rather than actions, and never states what the skill actually does with them. It is above anchor 3 (only 1-2 concrete actions) but the verb-less phrasing keeps it below anchor 5's comprehensive action coverage.

4 / 5

Completeness

The description is entirely a 'when' clause ('Use when working with ComfyUI workflows in OpenMontage, including...'); the 'what' is only implied by the enumerated topics, with no explicit statement of what the skill does. It is clearly above anchor 2 ('Use when working with documents' — vague in both dimensions) because the enumeration conveys concrete scope, but below anchor 4, which requires both 'what' and 'when' explicitly.

3 / 5

Trigger Term Quality

Strong natural terms for this domain: 'ComfyUI', 'comfyui_image/comfyui_video/comfyui_music', 'LoRAs', 'low-VRAM', 'community workflow imports'. However, common user phrasings such as 'generate an image/video/song with ComfyUI' and model family names a user would actually say (Wan, LTXV, ACE-Step) are missing, so it falls short of anchor 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

'ComfyUI workflows in OpenMontage, including comfyui_image/comfyui_video/comfyui_music' carves out a clear niche with distinct, tool-specific triggers; a user would only invoke this when working with these exact tools, so conflict risk is minimal.

5 / 5

Total

16

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

calesthio/OpenMontage

Path

.agents/skills/comfyui/SKILL.md

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.