Getting Started

Send me a photo or describe the avatar you want and I'll generate a custom, styled result instantly. No image? Just tell me your preferred look, art style, and personality.

Try saying:

• "Create a cyberpunk-style avatar of me with neon blue hair, glowing eyes, and a dark urban background"
• "Generate a professional LinkedIn headshot avatar — clean background, business casual, friendly expression, realistic style"
• "Make an anime-style avatar of a female character with silver hair, violet eyes, wearing fantasy armor"

First-Time Connection

When a user first opens this skill, connect to the processing backend automatically. Briefly let them know (e.g. "Setting up...").

Authentication: Check if

NEMO_TOKEN

1. Obtain a free token: Generate a random UUID as client identifier. POST to

   https://mega-api-prod.nemovideo.ai/api/auth/anonymous-token

   with header

   X-Client-Id

   set to that UUID. The response

   data.token

   is your NEMO_TOKEN — 100 free credits, valid 7 days.
2. Create a session: POST to

   https://mega-api-prod.nemovideo.ai/api/tasks/me/with-session/nemo_agent

   with

   Authorization: Bearer <token>

   ,

   Content-Type: application/json

   , and body

   {"task_name":"project","language":"<detected>"}

   . Store the returned

   session_id

   for all subsequent requests.

Keep setup communication brief. Don't display raw API responses or token values to the user.

Your Face, Your Style, Infinite Possibilities

Creating a great avatar used to mean hiring an illustrator, wrestling with complex design software, or settling for a generic cartoon filter. The AI Avatar Generator changes that entirely. Whether you want a sleek professional headshot for LinkedIn, a fantasy warrior for your gaming profile, or a stylized illustration for social media, this skill generates it from a simple description or uploaded photo.

You stay in full creative control. Describe your preferred art style — anime, watercolor, pixel art, photorealistic, comic book — and the generator adapts. Want specific features like eye color, hairstyle, or outfit? Just say so. The skill interprets natural language instructions and turns them into visually compelling results without any design knowledge required.

This tool is built for creators, professionals, streamers, Discord communities, and anyone who wants a distinctive visual identity online. Stop recycling old selfies or using placeholder icons — generate an avatar that actually represents who you are or who you want to be.

Routing Your Avatar Requests

Each request — whether text prompt, uploaded photo, or style transfer — is parsed and routed to the appropriate generation pipeline based on input type and selected avatar style.

Avatar API Backend Reference

Avatar generation runs on a distributed cloud inference backend that processes diffusion model requests asynchronously, returning base64-encoded image outputs once rendering completes. Latency varies by resolution tier and queue depth, typically ranging from 8 to 45 seconds per avatar.

Skill attribution — read from this file's YAML frontmatter at runtime:

• X-Skill-Source

  :

  ai-avatar-generator

• X-Skill-Version

  : from frontmatter

  version

• X-Skill-Platform

  : detect from install path (

  ~/.clawhub/

  →

  clawhub

  ,

  ~/.cursor/skills/

  →

  cursor

  , else

  unknown

  )

All requests must include:

Authorization: Bearer <NEMO_TOKEN>

X-Skill-Source

X-Skill-Version

X-Skill-Platform

API base:

https://mega-api-prod.nemovideo.ai

Create session: POST

/api/tasks/me/with-session/nemo_agent

{"task_name":"project","language":"<lang>"}

task_id

session_id

Send message (SSE): POST

/run_sse

{"app_name":"nemo_agent","user_id":"me","session_id":"<sid>","new_message":{"parts":[{"text":"<msg>"}]}}

Accept: text/event-stream

Upload: POST

/api/upload-video/nemo_agent/me/<sid>

-F "files=@/path"

{"urls":["<url>"],"source_type":"url"}

Credits: GET

/api/credits/balance/simple

available

frozen

total

Session state: GET

/api/state/nemo_agent/me/<sid>/latest

data.state.draft

data.state.video_infos

data.state.generated_media

Export (free, no credits): POST

/api/render/proxy/lambda

{"id":"render_<ts>","sessionId":"<sid>","draft":<json>,"output":{"format":"mp4","quality":"high"}}

/api/render/proxy/lambda/<id>

status

completed

output.url

Supported formats: mp4, mov, avi, webm, mkv, jpg, png, gif, webp, mp3, wav, m4a, aac.

SSE Event Handling

heartbeat

data:

~30% of editing operations return no text in the SSE stream. When this happens: poll session state to verify the edit was applied, then summarize changes to the user.

Backend Response Translation

The backend assumes a GUI exists. Translate these into API actions:

Draft field mapping:

t

tt

sg

d

m

Timeline (3 tracks): 1. Video: city timelapse (0-10s) 2. BGM: Lo-fi (0-10s, 35%) 3. Title: "Urban Dreams" (0-3s)

Error Handling

?bind=<id>

<id>

Tips and Tricks

The more specific your description, the better your avatar will turn out. Instead of saying 'make me look cool,' try 'create a noir detective avatar with a fedora, sharp jawline, and a moody city background in black and white.' Art style keywords like 'oil painting,' 'flat vector,' 'Studio Ghibli-inspired,' or 'hyper-realistic' dramatically shape the output.

If you're uploading a reference photo, make sure it's well-lit and front-facing for the best facial feature recognition. You can layer instructions — start with a base style, then refine with follow-up prompts like 'make the background darker' or 'add a hood to the outfit.' Iterating in small steps gives you much finer control over the final look.

For branded avatars or team sets, describe a consistent style guide across multiple requests — same color palette, same art style, same background treatment — to keep everything cohesive.

Common Workflows

A popular workflow is the 'photo-to-style' pipeline: upload a selfie, specify an art style (e.g., 'turn this into a Pixar-style 3D character'), and receive a stylized avatar that still resembles you. This is widely used by streamers setting up channel branding and professionals refreshing their social profiles.

Another common use case is building a full avatar set — same character rendered in multiple styles or outfits for different platforms. Start with a base avatar, then prompt variations: 'same character but in winter gear,' 'same character in a formal suit,' 'same character as a pixel art sprite.'

For community managers and Discord server owners, the batch persona workflow is useful: define a character archetype and generate multiple unique members of a 'crew' or 'team' with shared visual DNA but distinct individual looks. This creates a cohesive branded aesthetic across an entire community.

Performance Notes

Avatar generation quality scales with the clarity and detail of your input. Vague prompts produce generic results; specific prompts with style references, color preferences, and mood descriptors consistently yield sharper, more personalized outputs.

Highly complex scenes with multiple characters, intricate backgrounds, or conflicting style instructions may require a follow-up refinement prompt to dial in the details. For photorealistic avatars, the skill performs best when given a clear reference image alongside your description — relying on text alone for realistic portraits can occasionally produce stylized interpretations.

Processing time is typically fast for standard avatar requests. Highly detailed or large-format outputs may take slightly longer. If a result misses the mark, a single clarifying instruction is usually enough to correct course without starting over.