Skip to main content

langchain-anyapi

LangChain tools for AnyAPI: hundreds of data and scraping APIs (Reddit, Instagram, TikTok, YouTube, Google search, web scraping, and more) behind one key, billed per request in USD, no subscriptions.

pip install langchain-anyapi

Set ANYAPI_API_KEY, or pass api_key= to a tool or the toolkit. A key with starter credit is free at https://getanyapi.com.

The whole loop in one line

from langchain.agents import create_agent
from langchain_anyapi import ANYAPI_INSTRUCTIONS, AnyAPIToolkit

agent = create_agent(
    "anthropic:claude-sonnet-4-5",
    tools=AnyAPIToolkit().get_tools(),
    system_prompt=ANYAPI_INSTRUCTIONS,
)

Or take one tool at a time:

from langchain_anyapi import AnyAPIGetAPI, AnyAPIRunAPI, AnyAPISearchAPIs

AnyAPISearchAPIs().invoke({"query": "reddit trending posts"})
AnyAPIGetAPI().invoke({"sku_id": "reddit.trending_posts"})
AnyAPIRunAPI().invoke({"sku_id": "reddit.trending_posts", "input": {"limit": 2}})

Five tools, not one per API

Tool Charges? What it does
anyapi_search_apis no ranked search over the catalog, descriptions kept, schemas omitted
anyapi_list_apis no browse summaries, optionally filtered by category
anyapi_get_api no one API in full: strict input schema, output schema, per-lane USD pricing, 30-day latency
anyapi_run_api YES execute one API with a normalized input payload
anyapi_get_balance no remaining wallet balance in USD

AnyAPI publishes hundreds of APIs. One tool per API would not fit an agent's context, so these five teach the loop instead: search or list to find an API, anyapi_get_api to read its strict input schema, then anyapi_run_api. Input schemas reject unknown fields rather than ignoring them, so an input built from a description instead of a schema usually fails.

Money

Prices are USD and come off the wire exactly as the gateway published them. Every static price is quoted in both denominations: maxUsd is what one request is billed, and maxPer1kUsd is the same price per 1,000 requests. Nothing in this package multiplies one into the other. A run reports its actual charge as costUsd.

There is no quote tool here. The AnyAPI MCP server has one, but the published getanyapi SDK exposes no quote method, and adding one would mean hand-rolling an authenticated HTTP call beside the SDK. anyapi_get_api already publishes pricing.from.maxUsd, the most a first-choice run is billed, and pricing.failoverMaxUsd, the ceiling for any run, so an agent can bound its spend before it calls. One further difference from the MCP server's tools: search results carry no heavy marker.

Behavior worth knowing

  • Both _run and _arun are real. The async path uses AsyncAnyAPI, not the sync client on a thread.
  • The client is built on first use. Importing this package and constructing a tool need neither a key nor a socket, so a missing key surfaces only when a tool is actually called.
  • A failed call returns a readable payload with error, status, and where the gateway sent them code and requestId, rather than aborting the run.
  • response_format is "content". LangChain serializes the returned dict to JSON for the tool message either way, so an artifact would only carry a second copy of the same payload.
  • anyapi_run_api takes fields, max_items, and summary to trim a large response. None of them change what you are charged.

Development

pip install -e ".[dev]"
ruff check . && ruff format --check . && mypy && pytest

Unit tests run with sockets blocked and no key. The integration tests call the live gateway and are skipped unless ANYAPI_API_KEY is set; they cover the free tools only, because the standard suite invokes a tool repeatedly and every run of an API charges the wallet.

Links

MIT licensed.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

langchain_anyapi-0.2.0.tar.gz (17.1 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

langchain_anyapi-0.2.0-py3-none-any.whl (13.1 kB view details)

Uploaded Python 3

File details

Details for the file langchain_anyapi-0.2.0.tar.gz.

File metadata

  • Download URL: langchain_anyapi-0.2.0.tar.gz
  • Upload date:
  • Size: 17.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for langchain_anyapi-0.2.0.tar.gz
Algorithm Hash digest
SHA256 9f68aec6234cb39ed3c1af29ac52a44c73e619b9e2c647041834c7572327c7ff
MD5 ddf96b8543fbe9d435495ed434db973b
BLAKE2b-256 43fe7f2262a000b6685b3bbf7ed4e379c1d586e28ef202a5d66b618647503865

See more details on using hashes here.

Provenance

The following attestation bundles were made for langchain_anyapi-0.2.0.tar.gz:

Publisher: release-langchain-anyapi.yml on getanyapi-com/integrations

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file langchain_anyapi-0.2.0-py3-none-any.whl.

File metadata

File hashes

Hashes for langchain_anyapi-0.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 ced97b0a8f3e85808dcfee482bd4404f5e6633a3d17ddc970e14ef139e555317
MD5 12de66a3c614e8bfb90b0837b2d10796
BLAKE2b-256 824904161c2a35b7dbfa41489e857e30dc42ab3fa86f49d8a5c5439d8f9bae68

See more details on using hashes here.

Provenance

The following attestation bundles were made for langchain_anyapi-0.2.0-py3-none-any.whl:

Publisher: release-langchain-anyapi.yml on getanyapi-com/integrations

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

This release

0.2.0 This release

2 files

0.1.0

2 files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page