cracked-ai-pydantic-ai
Pydantic AI tools for Cracked, the tool router for AI agents: one API key, 67,000+ tools (web search, scrapers, social profiles, enrichment, media models) with measured health and a real price per call.
load_cracked_tools() turns every Cracked capability into a native Pydantic AI Tool, built at runtime from
GET https://cracked.ai/v1/capabilities/tools?format=openai via Tool.from_schema(). Runs go through POST /v1/run;
each result carries the run permalink, status and cost.
pip install cracked-ai-pydantic-ai pydantic-ai
import os
from pydantic_ai import Agent
from cracked_ai_pydantic_ai import load_cracked_tools
tools = load_cracked_tools(api_key=os.environ["CRACKED_API_KEY"], capabilities=["web-search", "weather-forecast"])
agent = Agent("openai:gpt-4o-mini", tools=tools, instructions="Use the tools; cite the run url.")
print(agent.run_sync("What is the weather in Austin this weekend?").output)
# Each tool call returns {"run": {"runId", "status", "url", "costUsd", "durationMs"}, "output": ...}
Options: capabilities (ids, dashes or underscores), set="top"|"all", base_url, timeout, max_output_chars.
Every tool returns {"run": {"runId", "status", "url", "costUsd", "durationMs"}, "output": ...}; API errors come back as
{"error", "status", "message"} so the model can read them.
Pricing
There is no subscription: each run is billed per call from your workspace balance. The price is shown on every
tool page (https://cracked.ai/tools) and inside each tool description your model sees (Price: from $.../call ...).
run.costUsd in every result is what that call actually cost. Full list: https://cracked.ai/capabilities and https://cracked.ai/pricing.
Keys, trial credit and referrals
- Humans: https://cracked.ai/app/keys.
- Agents can self-register without a human:
POST https://cracked.ai/v1/agents/register {"name": "my-agent"}returns anapiKey(shown once, with $1 of trial credit) and aclaimUrlto hand to your human; claiming adds credit and enables top-ups. - Every workspace that signs up through your referral link (
GET https://cracked.ai/v1/referrals) earns your workspace $5 of credit. - Try a tool with no key or install:
npx cracked-ai try "weather in austin".
Docs: https://cracked.ai/docs/integrations. Support: support@cracked.ai. MIT.
Metadata
Release files for cracked-ai-pydantic-ai 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 |
|---|---|---|---|---|
| cracked_ai_pydantic_ai-0.1.0.tar.gz | 6.4 kB | Without provenance | Details for cracked_ai_pydantic_ai-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| cracked_ai_pydantic_ai-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for cracked_ai_pydantic_ai-0.1.0-py3-none-any.whl |
Total release size: 14.2 kB
Release files / cracked_ai_pydantic_ai-0.1.0.tar.gz
| Download URL | cracked_ai_pydantic_ai-0.1.0.tar.gz |
|---|---|
| Size | 6.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
3f8d7c52d2c1ce4a55ac3435e5902c0abffb72d3c17c00552993cdc4dfc1bc61
|
|
BLAKE2b-256 checksum How to use checksums |
cd6279870942f68433cc5c88c50eae0e2bf6ec3e9f1ee5627fc86790bc3fe10c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.12.3
|
Release files / cracked_ai_pydantic_ai-0.1.0-py3-none-any.whl
| Download URL | cracked_ai_pydantic_ai-0.1.0-py3-none-any.whl |
|---|---|
| Size | 7.8 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
bbf90554d6242413089de607f49659b0ab49ae9c9810a841cdeb0f574da9cb9f
|
|
BLAKE2b-256 checksum How to use checksums |
ee64fcbb50d6557e80cc97196655ae910790156d2800712276dff43ba4ce353d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.12.3
|
Security for cracked-ai-pydantic-ai 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.