muapi-langchain
LangChain integration for muapi.ai — 250+ generative media models behind a single key, exposed as LangChain tools, a document loader, and a Deep Agents recipe.
Install
pip install muapi-langchain
# For the Deep Agents demo:
pip install "muapi-langchain[deepagents]"
Or install directly from the repo (latest main):
pip install "git+https://github.com/SamurAIGPT/muapi-cli.git#subdirectory=integrations/langchain"
Set your API key:
export MUAPI_API_KEY="..." # or: muapi auth configure
What's in the box
4 tools (capability gradient)
| Tool | What it does | Costs credits? |
|---|---|---|
muapi_select |
Rank models + skills for an intent | No |
muapi_generate |
Single-shot generation (image / video / audio / edit) | Yes (per call) |
muapi_run_skill |
Named multi-step recipe (UGC ad, storyboard, …) | Yes (recipe) |
muapi_creative_agent |
Open-ended brief → planner → executor | Yes (variable) |
from muapi_langchain import muapi_select, muapi_generate
print(muapi_select.invoke({
"intent": "cinematic product photo of a sneaker",
"kind": "image", "tier": "best", "limit": 3,
}))
print(muapi_generate.invoke({
"prompt": "A glossy sneaker on a wet street at neon-lit night",
"kind": "image", "tier": "best",
}))
MuapiAssetLoader — document loader
Hydrate prior muapi generations as Documents for RAG / eval.
from muapi_langchain import MuapiAssetLoader
docs = MuapiAssetLoader(request_ids=["req_abc", "req_def"]).load()
for d in docs:
print(d.metadata["url"], "·", d.page_content[:80])
MuapiCostCallback — budget tracking
Pipe every muapi tool call through a callback that accumulates credit spend and aborts the agent when a budget cap is hit.
from muapi_langchain import MuapiCostCallback
cost_cb = MuapiCostCallback(
budget_credits=500,
on_event=lambda evt, payload: print(evt, payload),
)
agent.invoke({...}, config={"callbacks": [cost_cb]})
print(cost_cb.summary())
Deep Agents recipe
The recommended pattern: cheap tools (select, generate) on the main
planner, heavy tools (run_skill, creative_agent) on a
creative-specialist subagent, with interrupt_on for human approval of
open-ended creative work.
See examples/deep_agents_demo.py for the
full runnable example.
from deepagents import create_deep_agent
from muapi_langchain import PLANNER_TOOLS, SPECIALIST_TOOLS
agent = create_deep_agent(
model=...,
tools=PLANNER_TOOLS,
subagents=[{
"name": "creative-specialist",
"description": "Heavy muapi workflows and open-ended briefs.",
"system_prompt": "...",
"tools": SPECIALIST_TOOLS,
}],
interrupt_on={
"muapi_creative_agent": {"allowed_decisions": ["approve", "edit", "reject"]},
},
)
Decision tree
User brief
├─ Don't know which model/skill? → muapi_select (free)
├─ Single asset, clear prompt? → muapi_generate
├─ Matches a known recipe? → muapi_run_skill
└─ Multi-asset / multi-modal? → muapi_creative_agent (interrupt_on)
License
MIT.
Metadata
Release files for muapi-langchain 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| muapi_langchain-0.1.0.tar.gz | 53.8 kB | Without provenance | Details for muapi_langchain-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| muapi_langchain-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for muapi_langchain-0.1.0-py3-none-any.whl |
Total release size: 106.1 kB
Release files / muapi_langchain-0.1.0.tar.gz
| Download URL | muapi_langchain-0.1.0.tar.gz |
|---|---|
| Size | 53.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
a38f13fa3bddb5d12114b74409480c71b21ae863078ba9130a89cc845d8a362e
|
|
BLAKE2b-256 checksum How to use checksums |
109fc97c148b8a71fc7201eb0a9b0ba23b75a3580f7470bb58b20fbedf7c6b2b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.11.6
|
Release files / muapi_langchain-0.1.0-py3-none-any.whl
| Download URL | muapi_langchain-0.1.0-py3-none-any.whl |
|---|---|
| Size | 52.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
02f464bd325880170943e2db06ed441fb3f7ccccc271afa8db3ed81f0880fd74
|
|
BLAKE2b-256 checksum How to use checksums |
7c018def4d3bdf0bd1f40a5ab85b1e73cc1bf4e1cd3c1af8c3426d22dcd068e2
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.11.6
|
Security for muapi-langchain 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.