comfyui-prompt-engineer
Craft model-specific prompts optimized for the target checkpoint and identity method. Handles FLUX, SDXL, SD1.5, and Wan video models with proper syntax, quality tags, and negative prompts. Use when generating or refining prompts for ComfyUI workflows.
- Installs
- 1.3K
- Last 24h
- —
- Trending rank
- —
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/mckruz/comfyui-expert --skill comfyui-prompt-engineer- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads comfyui-prompt-engineer automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
ComfyUI Prompt Engineer
Generates optimized prompts tailored to specific models and identity methods. Different models respond differently to prompts.
Model-Specific Prompt Rules
FLUX.1 (dev/schnell/Kontext)
- Style: Natural language descriptions work best
- CFG: 3.5-4 (very low)
- Quality tags: Minimal - FLUX doesn't need "masterpiece, best quality"
- Length: Medium (50-100 words)
- Structure:
{subject description}, {setting}, {lighting}, {camera/style}
Good FLUX prompt:
photorealistic portrait of a woman with auburn hair and green eyes, freckles across
her nose and cheeks, wearing a cream knit sweater, sitting in a cafe with warm ambient
lighting, shallow depth of field, shot on Sony A7IV, 85mm lens
Bad FLUX prompt (too many quality tags):
masterpiece, best quality, 8k uhd, highly detailed, photorealistic portrait...
SDXL (RealVisXL, Juggernaut, etc.)
- Style: Quality tags at front help significantly
- CFG: 7-9
- Quality tags: Include
masterpiece, best quality, photorealistic - Length: Medium-long (50-150 words)
- Structure:
{quality tags}, {trigger word}, {subject}, {details}, {setting}, {style} - Weighted syntax: Supported
(important:1.3)or((very important))
Good SDXL prompt:
masterpiece, best quality, sage_character, photorealistic portrait of a woman,
detailed skin texture with freckles, emerald green eyes, auburn copper hair,
natural lighting from window, indoor setting, shallow depth of field,
RAW photo quality, 8k uhd, film grain
SD 1.5
- Style: Tag-based works best
- CFG: 7-8
- Quality tags: Essential
- Length: Shorter (30-80 words)
- Structure:
{quality}, {trigger}, {subject}, {details}, {style tags}
Wan 2.1/2.2 (Video)
- Style: Concise motion descriptions
- CFG: 5-7
- Quality tags: Minimal
- Length: Short (20-50 words)
- Focus: Describe the motion, not just the appearance
- Structure:
{subject}, {action/motion}, {setting}, {quality}
Good Wan prompt:
young woman with auburn hair, talking naturally with gentle hand gestures,
seated at a modern desk, soft studio lighting, high quality
AnimateDiff
- Style: Same as base model (SD1.5/SDXL) but add motion keywords
- Motion keywords: Include camera/subject motion description
- Length: Same as base model
With Identity Methods
InstantID Prompts
Key adjustments:
- DO NOT describe specific facial features (the model provides them)
- DO describe everything else: clothing, pose, setting, lighting
- Keep CFG at 4-5
photorealistic portrait, wearing black leather jacket, standing in an alley,
dramatic side lighting, urban setting, moody atmosphere, 8k quality
PuLID Prompts
Key adjustments:
- Can include some facial descriptions (PuLID is more tolerant)
- Method "neutral" for realistic, "fidelity" for exact match
- CFG 5-7
sage_character, photorealistic portrait, green eyes visible, natural makeup,
professional headshot, neutral grey background, studio lighting
IP-Adapter FaceID
Key adjustments:
- Describe the style you want, not the face
- weight_type "style transfer" for 3D→realistic
- Standard CFG for base model
photorealistic portrait, DSLR photo quality, natural skin texture,
warm indoor lighting, bokeh background, professional photography
FLUX Kontext (Editing)
Key adjustments:
- Describe the EDIT, not the full image
- Be specific about what to change
- Mention what to preserve
Change the outfit to a formal black evening dress while keeping the face,
hair, and pose exactly the same. Add subtle jewelry.
With Character LoRA
Key adjustments:
- ALWAYS include trigger word first
- Don't describe features the LoRA has learned
- Focus on what varies: pose, clothing, setting, lighting
sage_character, standing on a rooftop at sunset, wind blowing hair,
wearing casual summer dress, city skyline behind, golden hour lighting,
cinematic composition
Negative Prompt Templates
Universal Negative (SDXL/SD1.5)
(worst quality:1.4), (low quality:1.4), blurry, deformed, bad anatomy,
bad hands, extra fingers, missing fingers, extra limbs, fused fingers,
text, watermark, signature, jpeg artifacts, username, error
Photorealism Negative
3d render, cartoon, anime, illustration, painting, drawing, cgi,
plastic skin, smooth skin, airbrushed, video game, doll, mannequin,
oversaturated, artificial lighting
Video Negative (Wan/AnimateDiff)
static, frozen, jerky motion, low quality, blurry, distorted face,
bad anatomy, glitch, artifacts, flickering, jittery, unnatural movement
FLUX Negative (keep minimal)
blurry, low quality, distorted, deformed, ugly, watermark, text
Prompt Construction Workflow
- Get character profile from
projects/{project}/characters/{name}/profile.yaml - Get target model from inventory/user preference
- Apply model rules from above
- Add identity method modifiers if applicable
- Include trigger word if LoRA is being used
- Draft positive + negative pair
- Review against past successes in character's
generation_history - Adjust CFG recommendation based on method stack
Reference
references/prompt-templates.md- Full template library with examplesreferences/workflows.md- CFG and sampler settings per workflow- Character profiles in
projects/for trigger words and feature descriptions
FAQ
What does the comfyui-prompt-engineer skill do?
Craft model-specific prompts optimized for the target checkpoint and identity method. Handles FLUX, SDXL, SD1.5, and Wan video models with proper syntax, quality tags, and negative prompts. Use when generating or refining prompts for ComfyUI workflows.
How do I install comfyui-prompt-engineer?
Run npx skills add https://github.com/mckruz/comfyui-expert --skill comfyui-prompt-engineer in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/comfyui-prompt-engineer/ (personal) or .claude/skills/comfyui-prompt-engineer/ (project).
Which AI agents can use comfyui-prompt-engineer?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is comfyui-prompt-engineer safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains comfyui-prompt-engineer?
It is published by mckruz in mckruz/comfyui-expert. It has 1,307 recorded installs.
More from mckruz/comfyui-expert
comfyui-workflow-builder
mckruz/comfyui-expert
Generate, build, create, or design ComfyUI workflow JSON from natural language descriptions. Produces valid node graphs with correct classtypes, connections, output indices, and model-appropriate settings. Handles txt2img, img2img, inpainting, ControlNet, LoRA stacking, upscaling, and face detailing pipelines. Does NOT cover ComfyUI installation, custom node development, Python scripting, model training, hardware advice, or architectural explanations.
comfyui-api
mckruz/comfyui-expert
Connect to a running ComfyUI instance, queue workflows, monitor execution, and retrieve results. Supports both online (REST API) and offline (JSON export) modes. Use when executing ComfyUI workflows or checking server status.
comfyui-video-pipeline
mckruz/comfyui-expert
Generate videos using ComfyUI with Wan 2.2, FramePack, or AnimateDiff. Handles image-to-video, text-to-video, talking heads, and motion-controlled animation. Use when creating any video content from character images or text descriptions.
comfyui-troubleshooter
mckruz/comfyui-expert
Diagnose ComfyUI errors, workflow failures, and quality issues. Suggests fixes based on error patterns, missing dependencies, and community-known workarounds. Use when ComfyUI workflows fail or produce unexpected results.
comfyui-character-gen
mckruz/comfyui-expert
Build identity-preserving character generation workflows and pipelines in ComfyUI. Selects the optimal identity method (InfiniteYou, FLUX Kontext, PuLID, InstantID, IP-Adapter) based on use case requirements. Handles face preservation, likeness transfer, cross-domain conversion (3D to photo), multi-reference consistency, iterative character editing, and character variation generation. Triggers on requests to generate consistent characters, preserve identity across images, create face-swapping workflows, or convert 3D renders to photorealistic portraits. Does NOT cover general image generation without identity preservation, model training/LoRA fine-tuning, animation, technical explanations, or workflow debugging.
comfyui-inventory
mckruz/comfyui-expert
Discover and cache all installed ComfyUI models, custom nodes, and system capabilities. Works online (API queries) and offline (directory scanning). Use before generating workflows to verify available resources.
Popular ai & agents skills
google-agents-cli-publish
google/agents-cli
This skill should be used when the user wants to "publish an agent", "publish my ADK agent", "register an agent with Gemini", "publish to Gemini" (formerly Gemini Enterprise), or needs guidance on the agents-cli publish gemini command. Also use when the user wants to "manage agents in Agent Registry", "list/update/delete registered agents", or "register an MCP server". Covers ADK vs A2A registration modes, programmatic and interactive usage, flag reference, auto-detection from deployment metadata, Agent Registry fleet management, and troubleshooting. Part of the agents-cli skills suite. Do NOT use for deployment (use google-agents-cli-deploy).
google-agents-cli-eval
google/agents-cli
This skill should be used when the user wants to "run an evaluation", "evaluate my agent", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval results", "optimize agent", or needs guidance on the Gemini platform eval methodology and the Quality Flywheel. Covers eval metrics, dataset schema, LLM-as-judge scoring, and common failure causes. Applies to any agents-cli project, whatever framework the agent is written in. Do NOT use for agent API code patterns (ADK: use google-agents-cli-adk-code), deployment (use google-agents-cli-deploy), or project scaffolding (use google-agents-cli-scaffold).
skill-creator
anthropics/skills
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
codex-pet
prime-skills/runcomfy-agent-skills
Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into ${CODEXHOME:-$HOME/.codex}/pets/ / and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as runcomfy run openai/gpt-image-2/edit to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no $imagegen, no OPENAIAPIKEY required, only RUNCOMFYTOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.
full-output-enforcement
leonxlnx/taste-skill
Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.
executing-plans
obra/superpowers
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available