Meshy 3d Agent

Meshy 3D — Generation + Printing

Directly communicate with the Meshy AI API to generate and print 3D assets. Covers the complete lifecycle: API key setup, task creation, exponential backoff polling, downloading, multi-step pipelines, and 3D print preparation with slicer integration.

SECURITY MANIFEST

Environment variables accessed:

• MESHY_API_KEY — API authentication token sent in HTTP Authorization: Bearer header only. Never logged, never written to any file except .env in the current working directory when explicitly requested by the user.

External network endpoints:

• https://api.meshy.ai — Meshy AI API (task creation, status polling, model/image downloads)

File system access:

• Read: .env in the current working directory only (API key lookup)
• Write: .env in the current working directory only (API key storage, only on user request)
• Write: ./meshy_output/ in the current working directory (downloaded model files, metadata)
• Read: files explicitly provided by the user (e.g., local images passed for image-to-3D conversion), accessed only at the exact path the user specifies
• No access to home directories, shell profiles, or any path outside the above

Data leaving this machine:

• API requests to api.meshy.ai include the MESHY_API_KEY in the Authorization header and user-provided text prompts or image URLs. No other local data is transmitted. Downloaded model files are saved locally only.

IMPORTANT: First-Use Session Notice

When this skill is first activated in a session, inform the user:

All generated files will be saved to meshy_output/ in the current working directory. Each project gets its own folder ({YYYYMMDD_HHmmss}_{prompt}_{id}/) with model files, textures, thumbnails, and metadata. History is tracked in meshy_output/history.json.

This only needs to be said once per session.

IMPORTANT: File Organization

All downloaded files MUST go into a structured meshy_output/ directory in the current working directory. Do NOT scatter files randomly.

• Each project: meshy_output/{YYYYMMDD_HHmmss}_{prompt_slug}_{task_id_prefix}/
• Chained tasks (preview → refine → rig) reuse the same project_dir
• Track tasks in metadata.json per project, and global history.json
• Auto-download thumbnails alongside models

IMPORTANT: Shell Command Rules

Use only standard POSIX tools. Do NOT use rg, fd, bat, exa/eza.

IMPORTANT: Run Long Tasks Properly

Meshy generation takes 1–5 minutes. Write the entire create → poll → download flow as ONE Python script and execute in a single Bash call. Use python3 -u script.py for unbuffered output. Tasks sitting at 99% for 30–120s is normal finalization — do NOT interrupt.

Step 0: API Key Detection (ALWAYS RUN FIRST)

Only check the current session environment and the .env file in the current working directory. Do NOT scan home directories or shell profile files.

echo "=== Meshy API Key Detection ==="

# 1. Check current env var
if [ -n "$MESHY_API_KEY" ]; then
  echo "ENV_VAR: FOUND (${MESHY_API_KEY:0:8}...)"
else
  echo "ENV_VAR: NOT_FOUND"
fi

# 2. Check .env in current working directory only
if [ -f ".env" ] && grep -q "MESHY_API_KEY" ".env" 2>/dev/null; then
  echo "DOTENV(.env): FOUND"
  export MESHY_API_KEY=$(grep "^MESHY_API_KEY=" ".env" | head -1 | cut -d'=' -f2- | tr -d '"'"'" )
fi

# 3. Final status
if [ -n "$MESHY_API_KEY" ]; then
  echo "READY: key=${MESHY_API_KEY:0:8}..."
else
  echo "READY: NO_KEY_FOUND"
fi

# 4. Python requests check
python3 -c "import requests; print('PYTHON_REQUESTS: OK')" 2>/dev/null || echo "PYTHON_REQUESTS: MISSING (run: pip install requests)"

echo "=== Detection Complete ==="

Decision After Detection

• Key found → Proceed to Step 1.
• Key NOT found → Go to Step 0a.
• Python requests missing → Run pip install requests.

Step 0a: API Key Setup (Only If No Key Found)

Tell the user:

To use the Meshy API, you need an API key:

1. Go to https://www.meshy.ai/settings/api
2. Click "Create API Key", name it, and copy the key (starts with msy_)
3. The key is shown only once — save it somewhere safe

Note: API access requires a Pro plan or above. Free-tier accounts cannot create API keys.

Once the user provides the key, set it for the current session and optionally persist to .env:

# Set for current session only
export MESHY_API_KEY="msy_PASTE_KEY_HERE"

# Verify the key
STATUS=$(curl -s -o /dev/null -w "%{http_code}" \
  -H "Authorization: Bearer $MESHY_API_KEY" \
  https://api.meshy.ai/openapi/v1/balance)

if [ "$STATUS" = "200" ]; then
  BALANCE=$(curl -s -H "Authorization: Bearer $MESHY_API_KEY" https://api.meshy.ai/openapi/v1/balance)
  echo "Key valid. $BALANCE"
else
  echo "Key invalid (HTTP $STATUS). Please check the key and try again."
fi

To persist the key (current project only):

# Write to .env in current working directory
echo 'MESHY_API_KEY=msy_PASTE_KEY_HERE' >> .env
echo "Saved to .env"

# IMPORTANT: add .env to .gitignore to avoid leaking the key
grep -q "^\.env" .gitignore 2>/dev/null || echo ".env" >> .gitignore
echo ".env added to .gitignore"

Security reminder: The key is stored only in .env in your current project directory. Never commit this file to version control. .env has been automatically added to .gitignore.

Step 1: Confirm Plan With User Before Spending Credits

CRITICAL: Before creating any task, present the user with a cost summary and wait for confirmation:

I'll generate a 3D model of "<prompt>" using the following plan:

  1. Preview (mesh generation) — 20 credits
  2. Refine (texturing with PBR) — 10 credits
  3. Download as .glb

  Total cost: 30 credits
  Current balance: <N> credits

  Shall I proceed?

For multi-step pipelines (text-to-3d → rig → animate), show the FULL pipeline cost upfront.

Note: Rigging automatically includes walking + running animations at no extra cost. Only add Animate (3 credits) for custom animations beyond those.

Intent → API Mapping

Step 2: Execute the Workflow

Reusable Script Template

Use this as the base for ALL workflows. It loads the API key securely from environment or .env in the current directory only:

#!/usr/bin/env python3
"""Meshy API task runner. Handles create → poll → download."""
import requests, time, os, sys, re, json
from datetime import datetime

# --- Secure API key loading ---
def load_api_key():
    """Load MESHY_API_KEY from environment, then .env in cwd only."""
    key = os.environ.get("MESHY_API_KEY", "").strip()
    if key:
        return key
    env_path = os.path.join(os.getcwd(), ".env")
    if os.path.exists(env_path):
        with open(env_path) as f:
            for line in f:
                line = line.strip()
                if line.startswith("MESHY_API_KEY=") and not line.startswith("#"):
                    val = line.split("=", 1)[1].strip().strip('"').strip("'")
                    if val:
                        return val
    return ""

API_KEY = load_api_key()
if not API_KEY:
    sys.exit("ERROR: MESHY_API_KEY not set. Run Step 0a to configure it.")

# Never log the full key — only first 8 chars for traceability
print(f"API key loaded: {API_KEY[:8]}...")

BASE = "https://api.meshy.ai"
HEADERS = {"Authorization": f"Bearer {API_KEY}"}
SESSION = requests.Session()
SESSION.trust_env = False  # bypass any system proxy settings

def create_task(endpoint, payload):
    resp = SESSION.post(f"{BASE}{endpoint}", headers=HEADERS, json=payload, timeout=30)
    if resp.status_code == 401:
        sys.exit("ERROR: Invalid API key (401). Re-run Step 0a.")
    if resp.status_code == 402:
        try:
            bal = SESSION.get(f"{BASE}/openapi/v1/balance", headers=HEADERS, timeout=10)
            balance = bal.json().get("balance", "unknown")
            sys.exit(f"ERROR: Insufficient credits (402). Balance: {balance}. Top up at https://www.meshy.ai/pricing")
        except Exception:
            sys.exit("ERROR: Insufficient credits (402). Check balance at https://www.meshy.ai/pricing")
    if resp.status_code == 429:
        sys.exit("ERROR: Rate limited (429). Wait and retry.")
    resp.raise_for_status()
    task_id = resp.json()["result"]
    print(f"TASK_CREATED: {task_id}")
    return task_id

def poll_task(endpoint, task_id, timeout=600):
    """Poll with exponential backoff (5s→30s, fixed 15s at 95%+)."""
    elapsed, delay, max_delay, backoff, finalize_delay, poll_count = 0, 5, 30, 1.5, 15, 0
    while elapsed < timeout:
        poll_count += 1
        resp = SESSION.get(f"{BASE}{endpoint}/{task_id}", headers=HEADERS, timeout=30)
        resp.raise_for_status()
        task = resp.json()
        status = task["status"]
        progress = task.get("progress", 0)
        bar = f"[{'█' * int(progress/5)}{'░' * (20 - int(progress/5))}] {progress}%"
        print(f"  {bar} — {status} ({elapsed}s, poll #{poll_count})", flush=True)
        if status == "SUCCEEDED":
            return task
        if status in ("FAILED", "CANCELED"):
            msg = task.get("task_error", {}).get("message", "Unknown")
            sys.exit(f"TASK_{status}: {msg}")
        current_delay = finalize_delay if progress >= 95 else delay
        time.sleep(current_delay)
        elapsed += current_delay
        if progress < 95:
            delay = min(delay * backoff, max_delay)
    sys.exit(f"TIMEOUT after {timeout}s ({poll_count} polls)")

def download(url, filepath):
    """Download a file into a project directory (within cwd/meshy_output/)."""
    os.makedirs(os.path.dirname(filepath), exist_ok=True)
    print(f"Downloading {filepath}...", flush=True)
    resp = SESSION.get(url, timeout=300, stream=True)
    resp.raise_for_status()
    with open(filepath, "wb") as f:
        for chunk in resp.iter_content(chunk_size=8192):
            f.write(chunk)
    print(f"DOWNLOADED: {filepath} ({os.path.getsize(filepath)/1024/1024:.1f} MB)")

# --- File organization helpers ---
OUTPUT_ROOT = os.path.join(os.getcwd(), "meshy_output")
os.makedirs(OUTPUT_ROOT, exist_ok=True)
HISTORY_FILE = os.path.join(OUTPUT_ROOT, "history.json")

def get_project_dir(task_id, prompt="", task_type="model"):
    slug = re.sub(r'[^a-z0-9]+', '-', (prompt or task_type).lower())[:30].strip('-')
    folder = f"{datetime.now().strftime('%Y%m%d_%H%M%S')}_{slug}_{task_id[:8]}"
    project_dir = os.path.join(OUTPUT_ROOT, folder)
    os.makedirs(project_dir, exist_ok=True)
    return project_dir

def record_task(project_dir, task_id, task_type, stage, prompt="", files=None):
    meta_path = os.path.join(project_dir, "metadata.json")
    meta = json.load(open(meta_path)) if os.path.exists(meta_path) else {
        "project_name": prompt or task_type, "folder": os.path.basename(project_dir),
        "root_task_id": task_id, "created_at": datetime.now().isoformat(), "tasks": []
    }
    meta["tasks"].append({"task_id": task_id, "task_type": task_type, "stage": stage,
                          "files": files or [], "created_at": datetime.now().isoformat()})
    meta["updated_at"] = datetime.now().isoformat()
    json.dump(meta, open(meta_path, "w"), indent=2)
    history = json.load(open(HISTORY_FILE)) if os.path.exists(HISTORY_FILE) else {"version": 1, "projects": []}
    folder = os.path.basename(project_dir)
    entry = next((p for p in history["projects"] if p["folder"] == folder), None)
    if entry:
        entry.update({"task_count": len(meta["tasks"]), "updated_at": meta["updated_at"]})
    else:
        history["projects"].append({"folder": folder, "prompt": prompt, "task_type": task_type,
            "root_task_id": task_id, "created_at": meta["created_at"],
            "updated_at": meta["updated_at"], "task_count": len(meta["tasks"])})
    json.dump(history, open(HISTORY_FILE, "w"), indent=2)

def save_thumbnail(project_dir, url):
    path = os.path.join(project_dir, "thumbnail.png")
    if os.path.exists(path): return
    try:
        r = SESSION.get(url, timeout=15); r.raise_for_status()
        open(path, "wb").write(r.content)
    except Exception: pass

Text to 3D (Preview + Refine)

Append to the template above:

PROMPT = "USER_PROMPT"

# Preview
preview_id = create_task("/openapi/v2/text-to-3d", {
    "mode": "preview",
    "prompt": PROMPT,
    "ai_model": "latest",
    # "pose_mode": "t-pose",   # Use "t-pose" if rigging/animating later
})
task = poll_task("/openapi/v2/text-to-3d", preview_id)
project_dir = get_project_dir(preview_id, prompt=PROMPT)
download(task["model_urls"]["glb"], os.path.join(project_dir, "preview.glb"))
record_task(project_dir, preview_id, "text-to-3d", "preview", prompt=PROMPT, files=["preview.glb"])
if task.get("thumbnail_url"):
    save_thumbnail(project_dir, task["thumbnail_url"])
print(f"\nPREVIEW COMPLETE — Task: {preview_id} | Project: {project_dir}")

# Refine
refine_id = create_task("/openapi/v2/text-to-3d", {
    "mode": "refine",
    "preview_task_id": preview_id,
    "enable_pbr": True,
    "ai_model": "latest",
})
task = poll_task("/openapi/v2/text-to-3d", refine_id)
download(task["model_urls"]["glb"], os.path.join(project_dir, "refined.glb"))
record_task(project_dir, refine_id, "text-to-3d", "refined", prompt=PROMPT, files=["refined.glb"])
print(f"\nREFINE COMPLETE — Task: {refine_id} | Formats: {', '.join(task['model_urls'].keys())}")

Note: Only previews from meshy-5 or latest support refine. meshy-6 previews do NOT (API returns 400).

Image to 3D

import base64

# For local files: convert to data URI
# with open("photo.jpg", "rb") as f:
#     image_url = "data:image/jpeg;base64," + base64.b64encode(f.read()).decode()

task_id = create_task("/openapi/v1/image-to-3d", {
    "image_url": "IMAGE_URL_OR_DATA_URI",
    "should_texture": True,
    "enable_pbr": True,
    "ai_model": "latest",
})
task = poll_task("/openapi/v1/image-to-3d", task_id)
project_dir = get_project_dir(task_id, task_type="image-to-3d")
download(task["model_urls"]["glb"], os.path.join(project_dir, "model.glb"))
record_task(project_dir, task_id, "image-to-3d", "complete", files=["model.glb"])

Multi-Image to 3D

task_id = create_task("/openapi/v1/multi-image-to-3d", {
    "image_urls": ["URL_1", "URL_2", "URL_3"],  # 1–4 images
    "should_texture": True,
    "enable_pbr": True,
    "ai_model": "latest",
})
task = poll_task("/openapi/v1/multi-image-to-3d", task_id)
project_dir = get_project_dir(task_id, task_type="multi-image-to-3d")
download(task["model_urls"]["glb"], os.path.join(project_dir, "model.glb"))

Retexture

task_id = create_task("/openapi/v1/retexture", {
    "input_task_id": "PREVIOUS_TASK_ID",
    "text_style_prompt": "wooden texture",
    "enable_pbr": True,
})
task = poll_task("/openapi/v1/retexture", task_id)
project_dir = get_project_dir(task_id, task_type="retexture")
download(task["model_urls"]["glb"], os.path.join(project_dir, "retextured.glb"))

Remesh / Format Conversion

task_id = create_task("/openapi/v1/remesh", {
    "input_task_id": "TASK_ID",
    "target_formats": ["glb", "fbx", "obj"],
    "topology": "quad",
    "target_polycount": 10000,
})
task = poll_task("/openapi/v1/remesh", task_id)
project_dir = get_project_dir(task_id, task_type="remesh")
for fmt, url in task["model_urls"].items():
    download(url, os.path.join(project_dir, f"remeshed.{fmt}"))

Auto-Rigging + Animation

When the user asks to rig or animate, the generation step MUST use pose_mode: "t-pose".

# Pre-rig check: polycount must be ≤ 300,000
source_endpoint = "/openapi/v2/text-to-3d"  # adjust to match source task endpoint
source_task_id = "TASK_ID"
check = SESSION.get(f"{BASE}{source_endpoint}/{source_task_id}", headers=HEADERS, timeout=30)
check.raise_for_status()
face_count = check.json().get("face_count", 0)
if face_count > 300000:
    sys.exit(f"ERROR: {face_count:,} faces exceeds 300,000 limit. Remesh first.")

# Rig
rig_id = create_task("/openapi/v1/rigging", {
    "input_task_id": source_task_id,
    "height_meters": 1.7,
})
rig_task = poll_task("/openapi/v1/rigging", rig_id)
project_dir = get_project_dir(rig_id, task_type="rigging")
download(rig_task["result"]["rigged_character_glb_url"], os.path.join(project_dir, "rigged.glb"))
download(rig_task["result"]["basic_animations"]["walking_glb_url"], os.path.join(project_dir, "walking.glb"))
download(rig_task["result"]["basic_animations"]["running_glb_url"], os.path.join(project_dir, "running.glb"))

# Custom animation (optional, 3 credits — only if user needs beyond walking/running)
# anim_id = create_task("/openapi/v1/animations", {"rig_task_id": rig_id, "action_id": 1})
# anim_task = poll_task("/openapi/v1/animations", anim_id)
# download(anim_task["result"]["animation_glb_url"], os.path.join(project_dir, "animated.glb"))

Text to Image / Image to Image

# Text to Image
task_id = create_task("/openapi/v1/text-to-image", {
    "ai_model": "nano-banana-pro",
    "prompt": "a futuristic spaceship",
})
task = poll_task("/openapi/v1/text-to-image", task_id)
# Result URL: task["image_url"]

# Image to Image
task_id = create_task("/openapi/v1/image-to-image", {
    "ai_model": "nano-banana-pro",
    "prompt": "make it look cyberpunk",
    "reference_image_urls": ["URL"],
})
task = poll_task("/openapi/v1/image-to-image", task_id)

3D Printing Workflow

Trigger when the user mentions: print, 3d print, slicer, slice, bambu, orca, prusa, cura, figurine, miniature, statue, physical model, desk toy, phone stand.

Print Pipelines

Text-to-3D Print:

Image-to-3D Print:

Print Download + Slicer Script

Append to the template after task SUCCEEDED:

import subprocess, shutil

# Download OBJ for printing
obj_url = task["model_urls"].get("obj")
if not obj_url:
    print("OBJ not available. Available:", list(task["model_urls"].keys()))
    print("Download GLB and import manually into your slicer.")
    obj_url = task["model_urls"].get("glb")

obj_path = os.path.join(project_dir, "model.obj")
download(obj_url, obj_path)

# --- Post-process OBJ for slicer compatibility ---
def fix_obj_for_printing(input_path, output_path=None, target_height_mm=75.0):
    """
    Fix OBJ coordinate system, scale, and position for 3D printing slicers.
    - Rotates from glTF Y-up to slicer Z-up: (x, y, z) -> (x, -z, y)
    - Scales model to target_height_mm (default 75mm)
    - Centers model on XY plane (so slicer places it at bed center)
    - Aligns model bottom to Z=0 (origin at bottom)
    """
    if output_path is None:
        output_path = input_path

    lines = open(input_path, "r").readlines()

    # Pass 1: rotate vertices Y-up -> Z-up, collect bounds
    rotated = []
    min_x, max_x = float("inf"), float("-inf")
    min_y, max_y = float("inf"), float("-inf")
    min_z, max_z = float("inf"), float("-inf")
    for line in lines:
        if line.startswith("v "):
            parts = line.split()
            x, y, z = float(parts[1]), float(parts[2]), float(parts[3])
            rx, ry, rz = x, -z, y
            min_x, max_x = min(min_x, rx), max(max_x, rx)
            min_y, max_y = min(min_y, ry), max(max_y, ry)
            min_z, max_z = min(min_z, rz), max(max_z, rz)
            rotated.append(("v", rx, ry, rz, parts[4:]))
        elif line.startswith("vn "):
            parts = line.split()
            nx, ny, nz = float(parts[1]), float(parts[2]), float(parts[3])
            rotated.append(("vn", nx, -nz, ny, []))
        else:
            rotated.append(("line", line))

    model_height = max_z - min_z
    scale = target_height_mm / model_height if model_height > 1e-6 else 1.0
    x_offset = -(min_x + max_x) / 2.0 * scale
    y_offset = -(min_y + max_y) / 2.0 * scale
    z_offset = -(min_z * scale)

    # Pass 2: write transformed OBJ
    with open(output_path, "w") as f:
        for item in rotated:
            if item[0] == "v":
                _, rx, ry, rz, extra = item
                tx = rx * scale + x_offset
                ty = ry * scale + y_offset
                tz = rz * scale + z_offset
                extra_str = " " + " ".join(extra) if extra else ""
                f.write(f"v {tx:.6f} {ty:.6f} {tz:.6f}{extra_str}\n")
            elif item[0] == "vn":
                _, nx, ny, nz, _ = item
                f.write(f"vn {nx:.6f} {ny:.6f} {nz:.6f}\n")
            else:
                f.write(item[1])

    print(f"OBJ fixed: rotated Y-up→Z-up, scaled to {target_height_mm:.0f}mm, centered on XY, bottom at Z=0")

fix_obj_for_printing(obj_path, target_height_mm=75.0)
print(f"\nModel ready for printing: {os.path.abspath(obj_path)}")

target_height_mm: Default 75mm. Adjust based on user request (e.g. "print at 15cm" → 150.0).

Opening OBJ in slicer: When the user specifies a slicer (e.g. Bambu Studio, OrcaSlicer, Creality Print, PrusaSlicer, Cura), open the downloaded OBJ file directly:

• macOS: subprocess.run(["open", "-a", "<AppName>", obj_path]) — the OS resolves the app location automatically.
• Windows / Linux: Use shutil.which("<binary_name>") to find the executable in PATH, then subprocess.Popen([exe, obj_path]). If not found, print the file path and instruct manual open.
• No slicer specified: Print the OBJ file path and instruct: File → Import / Open → select .obj file.

Printability Checklist (Manual Review)

Automated printability analysis API is coming soon.

Multi-Color Printing (Manual Guidance)

Automated multi-color API is coming soon.

1. Use Retexture (10 credits) to apply distinct color regions
2. Download OBJ
3. In slicer's color painting tool, assign filament colors to regions
4. Slice with multi-color setup (Bambu AMS, Prusa MMU)

Step 3: Report Results

After task succeeds:

1. Downloaded file paths and sizes
2. Task IDs (for follow-up: refine, rig, retexture)
3. Available formats (list model_urls keys)
4. Credits consumed + current balance
5. Suggested next steps:
   • Preview done → "Want to refine (add textures)?"
   • Model done → "Want to rig this character?"
   • Rigged → "Want to apply a custom animation?"
   • Any model → "Want to 3D print this?"

Error Recovery

Task FAILED messages:

• "The server is busy..." → retry with backoff (5s, 10s, 20s)
• "Internal server error." → simplify prompt, retry once

Known Behaviors & Constraints

• 99% stall: Normal finalization (30–120s). Do NOT interrupt.
• Asset retention: Files deleted after 3 days (non-Enterprise). Download immediately.
• PBR maps: Must set enable_pbr: true explicitly.
• Refine: Only meshy-5 / latest previews support refine; meshy-6 does not.
• Rigging: Humanoid bipedal only, polycount ≤ 300,000.
• OBJ for printing: Always download OBJ for slicer compatibility (3MF not yet available from API). If user specifies a slicer, try to open OBJ directly; otherwise print file path for manual import.
• Timestamps: All API timestamps are Unix epoch milliseconds.

Execution Checklist

• Ran API key detection (Step 0) — checked env var and .env only
• API key verified (never printed in full)
• Presented cost summary and got user confirmation
• Wrote complete workflow as single Python script
• Ran with python3 -u for unbuffered output
• Reported file paths, formats, task IDs, and balance
• Suggested next steps

Additional Resources

For the complete API endpoint reference including all parameters, response schemas, and error codes, read reference.md.