langchain-cyberwarex
CyberWareX: web search, web page reading and token safety checks for AI agents, paid per call in USDC with x402. No account, no API key.
This package gives a LangChain or LangGraph agent two tools:
| Tool | What it does | Price per call (live quote 2026-11-06) |
|---|---|---|
CyberwarexSearch |
live web search, top results as title - url - snippet lines |
0.00435 USDC |
CyberwarexReadPage |
one web page as clean markdown, JavaScript rendered | 0.001 USDC |
Without a wallet the tools use the free trial (3 searches per day per IP). With a wallet key, each call is paid on its own over x402 in USDC on Base. The payment is gasless for the payer, so the wallet only needs USDC. There is no signup, no key to leak and no monthly plan.
Install
pip install langchain-cyberwarex # free trial only
pip install "langchain-cyberwarex[x402]" # pay per call from a wallet
Use
from langchain_cyberwarex import CyberwarexSearch, CyberwarexReadPage
tools = [CyberwarexSearch(), CyberwarexReadPage()] # free trial, no wallet
print(tools[0].invoke({"query": "x402 payment protocol"}))
# pay per call: pass a Base wallet key, or set CYBERWAREX_PRIVATE_KEY
tools = [CyberwarexSearch(private_key="0x..."), CyberwarexReadPage(private_key="0x...")]
Hand tools to any agent, for example create_agent(model, tools) in LangChain or create_react_agent in LangGraph.
Options: CyberwarexSearch(max_results=5, timeout=60), CyberwarexReadPage(max_chars=8000, timeout=90).
Errors (free trial used up, rate limit, network) come back as plain text the agent can read, not as a crash.
Measured
Run on 2026-11-06 with this package and a real paid call: search 4.7 s, page read 1.9 s. Our own logs put search at a median of about 2 s and a page read at about 5 s; slow pages take longer.
Use something else when
- You need more than a few thousand searches a day at a flat price: a monthly plan from a big search API costs less.
- You need guaranteed uptime with a contract: we are a small service with no SLA.
- Your agent cannot hold a wallet and the free trial is not enough.
- The page sits behind a login: the reader only sees public pages.
Safety
Page text is untrusted content from the web. The read tool tells the model never to follow instructions inside it, but treat it as untrusted in your own prompts too. Keep the wallet small: it only needs a few dollars of USDC.
More examples (TypeScript, CrewAI, Claude Code, Cursor): https://github.com/cyberwareX/x402-agent-search-examples Service: https://cyberwarex.com
MIT license.
Metadata
Release files for langchain-cyberwarex 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 |
|---|---|---|---|---|
| langchain_cyberwarex-0.1.0.tar.gz | 7.0 kB | Without provenance | Details for langchain_cyberwarex-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| langchain_cyberwarex-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for langchain_cyberwarex-0.1.0-py3-none-any.whl |
Total release size: 13.4 kB
Release files / langchain_cyberwarex-0.1.0.tar.gz
| Download URL | langchain_cyberwarex-0.1.0.tar.gz |
|---|---|
| Size | 7.0 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
8d4d3e0c1d242097f02b039a9e73057b989d2343aaf071868835114bd708f158
|
|
BLAKE2b-256 checksum How to use checksums |
1372e7755b12c84f46d5cdd0626616698dc0a018e043769352d1ce519bb27567
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.14.4
|
Release files / langchain_cyberwarex-0.1.0-py3-none-any.whl
| Download URL | langchain_cyberwarex-0.1.0-py3-none-any.whl |
|---|---|
| Size | 6.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
abacf97892d94bc98548cb7b4c883a410f1bee1a6c6b40e95554000137c052a3
|
|
BLAKE2b-256 checksum How to use checksums |
992f686f497b956c40cca8f34d128ad0b33e9ebb9667bd7b6cb4db35b590a554
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.14.4
|
Security for langchain-cyberwarex 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.