somanylemons 0.1.2

Python SDK for the SoManyLemons Public API — branded video reels, LinkedIn content, image quotes.

SoManyLemons Python SDK

Typed Python client for the SoManyLemons Public API. Submit recordings, render branded reels and image quotes, manage brands and drafts — from any Python app, no MCP required.

Install

pip install somanylemons

Quickstart

from somanylemons import SMLClient

client = SMLClient(api_key="sml_your_key_here")

# List recent recordings (compact metadata — no URLs)
for job in client.jobs.list(limit=5):
    print(f"{job.id} · {job.title} · {job.clip_count} clips")

# Submit a video and wait for rendered clips
job = client.reels.create_and_wait(
    url="https://example.com/recording.mp4",
    brand_profile_id=1,
    caption_style="LEMON",
    timeout=600,
)
for clip in job.clips:
    print(clip.url)

# Render an image quote with a pinned template (deterministic output)
result = client.image_quotes.create(
    quote_text="Labor productivity is the story of the next five years.",
    brand_profile_id=1,
    template_id=42,
)
print(result.image_url)

Configuration

Arg / env var	Default	Notes
api_key / SML_API_KEY	— (required)	Your sml_xxx key
base_url / SML_API_URL	https://api.somanylemons.com	Override for QAS / local
timeout	30.0	Seconds per request
max_retries	3	Retries on 5xx and 429

client = SMLClient(api_key="sml_xxx", base_url="https://api.qas.somanylemons.com")

Resources

Attribute	Endpoints
client.jobs	List, get, poll recordings
client.reels	Create reels (sync or blocking)
client.transcribe	Transcribe URL or file
client.image_quotes	Render branded quote images
client.brands	Manage brand profiles
client.drafts	Content queue
client.content	Generate / score / rewrite posts, extract quotes, usage

Error handling

from somanylemons import QuotaError, RateLimitError, JobFailedError, SMLClient

client = SMLClient(api_key="sml_xxx")

try:
    job = client.reels.create_and_wait(url="...")
except QuotaError as exc:
    print(f"Over quota: {exc.renders_used}/{exc.render_limit}")
except RateLimitError as exc:
    print(f"Rate limited, retry after {exc.retry_after}s")
except JobFailedError as exc:
    print(f"Render failed: {exc.message}")

Async

The async client (AsyncSMLClient) lands in 0.2.0. For now, wrap sync calls in asyncio.to_thread.

Security

The SDK is built so that common mistakes cannot leak your API key:

• Keys never appear in repr or logs. The client masks the key in its __repr__ and stores it in a _SecretStr wrapper. Explicit access requires .get().
• HTTPS enforced by default. Passing an http:// URL raises SMLError unless you opt in with allow_insecure_http=True (e.g. for local dev against localhost:8000).
• Foreign hosts rejected by default. Your base_url must be a SoManyLemons domain. If you need to point at a different host (a staging environment, a proxy), pass allow_foreign_host=True so the decision is explicit.
• Sensible retries only. 5xx and 429 responses retry with exponential backoff; 4xx errors (auth, permissions, validation) fail fast without retry so bad keys don't hammer the server.

If you're integrating the SDK in a shared / multi-tenant app, create one SMLClient per tenant so a per-request key override is never necessary.

Contributing

The project follows PEP 517/621 so any Python toolchain works (pip, uv, poetry, rye, pixi). No lockfile is published — consumers use their own.

# Clone and install in editable mode with dev dependencies
git clone https://github.com/NoMiddleInc/somanylemons-sdk.git
cd somanylemons-sdk
pip install -e ".[dev]"

# Run the test suite (pytest + respx for HTTP mocking)
pytest

# Lint and type-check
ruff check src tests
mypy src

Publish flow (maintainers):

pip install build twine
python -m build
twine upload dist/*

License

MIT