Skip to main content

langchain-breakwire

Current release: clients 0.7.0 (npm breakwire, PyPI breakwire, PyPI langchain-breakwire) · server 0.10.0 · GitHub Action breakwire-check 1.2.0 · CHANGELOG

LangChain integration for Breakwire: tools an agent can call to look up an API/SDK error or check upcoming deadlines, and a callback handler that looks up errors raised in your chains, LLM calls and tools.

pip install langchain-breakwire
from langchain_breakwire import breakwire_tools, BreakwireCallbackHandler

tools = breakwire_tools()      # lookup_error, deadlines, check_stack (one shared client); give them to your agent

handler = BreakwireCallbackHandler(service="openai-api")
try:
    chain.invoke(inputs, config={"callbacks": [handler]})
except Exception as e:
    print(e.breakwire.id, e.breakwire.fix)    # same exception, explanation attached
    raise
  • BreakwireLookupTool (breakwire_lookup_error): error text (+ optional service, sdk, sdk_version, model) → compact matches with id, confidence, fix, source URL. Unmatched errors are not logged unless log_miss=True.
  • BreakwireDeadlinesTool (breakwire_deadlines): upcoming act-by dates for a service.
  • BreakwireStackTool (breakwire_check_stack, 0.2.0): a package list or package.json / lockfiles / requirements.txt / pyproject.toml / workflow files (parsed locally since 0.6.0) → the deadlines, breaking changes and advisories that apply, soonest first.
  • BreakwireCallbackHandler: on_llm_error / on_tool_error / on_chain_error; each exception is looked up once even though LangChain reports it at every nesting level. Never raises, never changes the error. on_match(info, error) hook. 0.2.0: sends no_log=True by default (nothing stored server-side); no_log=False to contribute misses as gaps.
  • Tool results include remember: one short line the agent can save to memory (when to consult Breakwire again).

Built on the stdlib-only breakwire client, which strips secrets before sending. Breakwire's community verification has just started; claims link primary sources, prefer those.

MIT.

0.6.0

Requires breakwire>=0.6.0: the stack tool's files are parsed locally (only dependency and runtime facts are sent), redirects never carry credentials to another origin, and the callback handler forwards every lookup option (compact, min_score, max_matches, since, reflex, ...).

0.5.0

The default client sends X-Breakwire-Client: langchain-breakwire/<version> (aggregate per-channel counts on Breakwire 0.8.0). Requires breakwire>=0.5.0.

0.4.0

The lookup tool's compact result now includes cite, please and miss (gap id, coverage note, nearest claims). Requires breakwire>=0.4.0.

0.3.0

breakwire_tools() now returns five tools: breakwire_lookup_error, breakwire_close_lookup (every outcome), breakwire_changes (what changed since a date, with new model launches), breakwire_deadlines, breakwire_check_stack. All tools fail open: if Breakwire is unavailable they return {"error": ...} instead of raising.

Metadata

Release files for langchain-breakwire 0.7.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for langchain-breakwire 0.7.0
File Size Uploaded Provenance Actions
langchain_breakwire-0.7.0.tar.gz 10.2 kB With provenance Details for langchain_breakwire-0.7.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for langchain-breakwire 0.7.0
File Interpreter ABI Platform Actions
langchain_breakwire-0.7.0-py3-none-any.whl Python 3 none any Details for langchain_breakwire-0.7.0-py3-none-any.whl

Total release size: 18.5 kB

Release files / langchain_breakwire-0.7.0.tar.gz

Download URL langchain_breakwire-0.7.0.tar.gz
Size 10.2 kB
Tags Source
SHA-256 checksum
How to use checksums
e9527f798c274720c4b96d56f309032464b2f1a83f91f8450104090c21f2c763
BLAKE2b-256 checksum
How to use checksums
ef4da89639205faaaeb28543190ab6bfcf1f93b2eed0b670144b1d2e10b2fb75
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

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.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 29, 2026.

Transparency log

Release files / langchain_breakwire-0.7.0-py3-none-any.whl

Download URL langchain_breakwire-0.7.0-py3-none-any.whl
Size 8.3 kB
Tags Python 3
SHA-256 checksum
How to use checksums
2ef9535d04223d1b58fd2ebf7e52d19f2d4167f89157ae12423333807c27a4b6
BLAKE2b-256 checksum
How to use checksums
ebe261707a4627c71129c48fd3b520bf610f69065042ba7163fc139a58055f47
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

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.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 29, 2026.

Transparency log

Security for langchain-breakwire 0.7.0

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.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for langchain_breakwire-0.7.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 29, 2026.

Transparency log

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.

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