Clawf Selfie

Generate selfies by editing the current character's reference image via aibotclaw.com.

⚠️ CRITICAL: Always Use The Script

Never construct the API call inline. Always call the shell script. The script reads the reference image URL from character.json at runtime. If you hardcode any image URL, you will use the wrong image.

bash /root/.openclaw/workspace/skills/clawf-selfie/scripts/clawf-selfie.sh \
  "<PROMPT>" \
  "<CHANNEL>" \
  ["<CAPTION>"] \
  ["<IMAGE_SIZE>"]

When to Use

• User says "send a pic", "send me a pic", "send a photo", "send a selfie"
• User says "send a pic of you...", "send a selfie of you..."
• User asks "what are you doing?", "how are you doing?", "where are you?"
• User describes a context: "send a pic wearing...", "send a pic at..."

How to Use

Step 1: Select mode

Auto-select by keywords:

Step 2: Enrich the prompt

Do NOT pass the user's words directly. Expand into a vivid, detailed English description.

Use this checklist to enrich the <USER_CONTEXT> before inserting into the template:

Outfit style reference library — choose the closest match and expand:

Step 3: Assemble the final prompt

Character consistency note (Nano Banana model): Nano Banana has built-in character consistency when a reference image is provided via image_urls. Do NOT use repetitive anchor phrases like "same face, same body" — they are redundant and may confuse the model. Instead, describe what changes (outfit, scene, lighting) and leave everything else implicit. The model will preserve the character's face, body proportions, and breast size from the reference image automatically.

If a specific body feature is drifting across generations, describe it explicitly once in natural language (e.g. "her full figure", "her curvy silhouette") — not as a repeated keyword.

⚠️ MANDATORY consistency rules — apply to EVERY prompt without exception:

1. Human subject — the output must depict a human character, not an animal, pet, or object.
2. Strict face & body fidelity — the face, facial features (eyes, nose, lips, jawline), and body proportions MUST match the reference image exactly. Do NOT invent, alter, or idealize any facial feature or body shape. Any divergence from the reference is a hard failure.

Mirror mode — full-body, outfit showcase:

a photo of this exact person, same face and facial features as the reference image, same body proportions, but <ENRICHED_CONTEXT>. the person is taking a mirror selfie, natural relaxed pose, soft indoor lighting

Direct mode — close-up, location/scene:

a close-up selfie of this exact person, same face and facial features as the reference image, same body proportions, taken by herself at <ENRICHED_CONTEXT>, direct eye contact with the camera, looking straight into the lens, eyes centered and clearly visible, not a mirror selfie, phone held at arm's length, face fully visible

Step 4: Call the script

bash /root/.openclaw/workspace/skills/clawf-selfie/scripts/clawf-selfie.sh \
  "<FINAL_PROMPT>" \
  "<CHANNEL>"

The script will:

1. Read reference_image_url from /root/.openclaw/workspace/character.json
2. Call aibotclaw.com API with the correct reference image
3. Send the result to the target channel via OpenClaw

Step 5: Send (handled by the script automatically)

The script calls openclaw message send internally. No extra step needed.

Examples

# Mirror selfie — casual outfit (enriched)
bash /root/.openclaw/workspace/skills/clawf-selfie/scripts/clawf-selfie.sh \
  "a photo of this exact person, same face and facial features as the reference image, same body proportions, but wearing a loose dusty rose oversized knit sweater, light wash straight-leg jeans, white sneakers, hair in a casual messy bun, soft warm indoor lighting. the person is taking a mirror selfie, natural relaxed pose" \
  "telegram"

# Mirror selfie — sexy / going out
bash /root/.openclaw/workspace/skills/clawf-selfie/scripts/clawf-selfie.sh \
  "a photo of this exact person, same face and facial features as the reference image, same body proportions, but wearing a form-fitting black cut-out bodycon dress with a subtle plunging neckline, strappy heels, minimal gold jewelry, confident expression, soft studio lighting. the person is taking a mirror selfie, natural relaxed pose" \
  "telegram"

# Direct selfie — location scene
bash /root/.openclaw/workspace/skills/clawf-selfie/scripts/clawf-selfie.sh \
  "a close-up selfie of this exact person, same face and facial features as the reference image, same body proportions, taken by herself at a cozy Tokyo night market, warm amber street light, soft smile, slightly windswept hair, direct eye contact with the camera, looking straight into the lens, eyes centered and clearly visible, not a mirror selfie, phone held at arm's length, face fully visible" \
  "telegram"

# Direct selfie — beach swimwear
bash /root/.openclaw/workspace/skills/clawf-selfie/scripts/clawf-selfie.sh \
  "a close-up selfie of this exact person, same face and facial features as the reference image, same body proportions, taken by herself at a tropical beach, wearing a strappy sage green bikini, golden hour sunlight, sun-kissed glowing skin, relaxed candid smile, ocean waves blurred in background, direct eye contact with the camera, looking straight into the lens, not a mirror selfie, phone held at arm's length, face fully visible" \
  "telegram"

Supported Platforms

Error Handling

• CRS_API_KEY_ID missing: Set environment variable in openclaw.json
• 403 quota exhausted: Purchase more tokens at aibotclaw.com
• 504 timeout: Generation took >60s, retry is safe
• Wrong image: Never hardcode URLs — always call the script