guardrails-client

A thin REST client for the guardrails-api.

Quickstart

pip install guardrails-ai-sdk

from guardrails_ai.sdk import GuardrailsAI, Guard, ValidationOutcome

# Init the GuardrailsAI client
client = GuardrailsAI(api_key="xxx")

# Fetch a Guard from the server
guard: Guard = await client.guards.retrieve(name="my-guard")

print(guard)

# Run a Guard to validate content
validation_outcome: ValidationOutcome = await client.guards.validate(name="my-guard", llm_output="Hello, world.")

if not validation_outcome.validation_passed:
    print(validation_outcome.validation_summaries)

# Create Guarded Chat Completions
chat_completion = await client.guards.chat.completions.create(guard_name="my-guard", model="gpt-5-nano", messages=[{ "role": "user", "content": "Hello, world." }])

print(chat_completion.choices[0].message.content)
print(chat_completion.guardrails)

# Stream Guarded Chat Completions
completion_stream = await client.guards.chat.completions.create(
    guard_name="my-first-guard",
    model="gpt-5-nano",
    messages=[
        {"role": "user", "content": "Give yourself a realistic name and introduce yourself."}
    ],
    stream=True
)

full_text = ""
validation_summaries = []
async for chunk in completion_stream:
    full_text += chunk.choices[0].delta.content
    validation_summaries.extend(chunk.guardrails.get("validation_summaries", []))


print("\n ==> Validation Summaries: ", validation_summaries)
print("\n ==> Final Content: ", full_text)

Download files

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

Source Distribution

guardrails_ai_sdk-0.5.0.tar.gz (14.5 kB view details)

Uploaded Jul 10, 2026 Source

Built Distribution

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

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

guardrails_ai_sdk-0.5.0-py3-none-any.whl (8.8 kB view details)

Uploaded Jul 10, 2026 Python 3

File details

Details for the file guardrails_ai_sdk-0.5.0.tar.gz.

File metadata

Download URL: guardrails_ai_sdk-0.5.0.tar.gz
Upload date: Jul 10, 2026
Size: 14.5 kB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.13

File hashes

Hashes for guardrails_ai_sdk-0.5.0.tar.gz
Algorithm	Hash digest
SHA256	`366f9aa4136a588505083561451761d44ab8ae4a94a3d2b019f22215d06c8e2f`
MD5	`cff2d5b3c79259d7d71822568e4c33ab`
BLAKE2b-256	`3d00289ced2d93e61f86db14824e8b533cc425ea6c3ff0f168204d91950dce43`

See more details on using hashes here.

Provenance

The following attestation bundles were made for guardrails_ai_sdk-0.5.0.tar.gz:

Publisher: sdk_py_publish.yml on guardrails-ai/guardrails-ai

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

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: guardrails_ai_sdk-0.5.0.tar.gz
- Subject digest: 366f9aa4136a588505083561451761d44ab8ae4a94a3d2b019f22215d06c8e2f
- Sigstore transparency entry: 2133637159
- Sigstore integration time: Jul 10, 2026
Source repository:
- Permalink: guardrails-ai/guardrails-ai@31bac923cfbb4e2191e7d60ec22033a12571d668
- Branch / Tag: refs/heads/main
- Owner: https://github.com/guardrails-ai
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: sdk_py_publish.yml@31bac923cfbb4e2191e7d60ec22033a12571d668
- Trigger Event: push