Skip to main content

Python SDK for CRP-Comply — EU AI Act & GDPR compliance automation (CRPv5 / agentic positioning)

Project description

crp-comply-sdk

PyPI Python License

Python SDK for CRP-Comply — EU AI Act & GDPR compliance automation. A thin, typed HTTP client covering every user-facing endpoint of the CRP-Comply REST API, including the local LLM worker relay for private, on-premise inference.

Install

pip install crp-comply-sdk

# If you are running the local LLM worker relay, install the worker extra:
pip install 'crp-comply-sdk[worker]'

Quick Start

from crp_comply_sdk import CRPComply

with CRPComply(api_key="crp_...") as client:
    # Account
    me = client.me()
    usage = client.usage()

    # EU AI Act Article 6 risk classification
    risk = client.risk_assessment(
        system_name="Hiring Assistant",
        category="employment",
        affects_fundamental_rights=True,
    )

    # GDPR Article 35 DPIA
    dpia = client.dpia(
        system_name="Hiring Assistant",
        data_subjects=["candidates", "employees"],
        makes_automated_decisions=True,
    )

    # Full conformity evidence pack (zip with manifest + HMAC)
    pack = client.evidence_pack(system_name="Hiring Assistant", category="employment")
    zip_bytes = client.download_evidence_pack(pack["pack_id"])
    with open("evidence.zip", "wb") as f:
        f.write(zip_bytes)

API Surface

All methods are keyword-only. Every call is tier-gated server-side — a CRPComplyTierError is raised with an upgrade_url when a feature isn't available on your plan.

Compliance generation

Method Endpoint Purpose
risk_assessment(...) POST /risk-assessment EU AI Act Article 6
compliance_report(..., markdown=False) POST /compliance-report[/markdown] Full compliance status
dpia(...) POST /dpia GDPR Article 35 DPIA
transparency(system_name=...) POST /transparency AI Act Article 13
technical_docs(system_name=...) POST /technical-docs AI Act Article 11
audit_session(session_file=...) POST /audit Audit a CRP session
full_report(...) POST /full-report Markdown mega-report

Evidence packs

Method Endpoint
evidence_pack(...) POST /evidence-pack — build new pack
list_evidence_packs() GET /evidence-packs
get_evidence_pack(pack_id) GET /evidence-packs/{id} — manifest
download_evidence_pack(pack_id) GET /evidence-packs/{id}/download — raw zip bytes
delete_evidence_pack(pack_id) DELETE /evidence-packs/{id}

Persisted reports

Method Endpoint
list_reports(kind=None) GET /reports?kind=...
get_report(report_id) GET /reports/{id}
get_report_markdown(report_id) GET /reports/{id}/markdown — returns str
delete_report(report_id) DELETE /reports/{id}

SDK gateway (realtime checks)

Method Endpoint
features() GET /sdk/features — tier feature matrix
audit(prompt=..., response=...) POST /sdk/audit — PII + injection + risk
classify_risk(...) POST /sdk/classify — quick AI Act bucket

Account

Method Endpoint
health() GET /health (unauthenticated)
me() GET /me — profile + tier + provider status
usage() GET /usage — monthly quota breakdown

Configuration

Option Env var Default
api_key CRP_COMPLY_API_KEY — (required)
base_url CRP_COMPLY_BASE_URL https://comply.crprotocol.io/api/v1
timeout 60 seconds
import os
os.environ["CRP_COMPLY_API_KEY"] = "crp_..."
os.environ["CRP_COMPLY_BASE_URL"] = "https://comply.crprotocol.io/api/v1"

from crp_comply_sdk import CRPComply
client = CRPComply()  # picks up env vars

LLM Backends

CRP-Comply is LLM-agnostic. Configure your preferred backend server-side at Settings → LLM Provider. Supported:

Backend Default URL Notes
OpenAI https://api.openai.com/v1 Cloud
Anthropic https://api.anthropic.com/v1 Cloud
LM Studio http://localhost:1234/v1 Local, OpenAI-compatible
Ollama http://localhost:11434/v1 Local, OpenAI-compatible
Custom any Any OpenAI-compatible endpoint

Exception Hierarchy

CRPComplyError              # base — all SDK errors
├── CRPComplyAuthError      # 401, 403 — invalid / missing API key
├── CRPComplyQuotaError     # 429 — monthly quota exhausted (has .upgrade_url)
├── CRPComplyTierError      # 402 — feature not in tier (has .upgrade_url,
│                           #       .feature, .current_tier, .required_tier)
└── CRPComplyServerError    # 5xx — server-side failure
from crp_comply_sdk import CRPComply, CRPComplyTierError, CRPComplyQuotaError

try:
    pack = client.evidence_pack(system_name="X", category="y")
except CRPComplyTierError as exc:
    print(f"Upgrade required: {exc.upgrade_url}")
except CRPComplyQuotaError as exc:
    print(f"Quota exhausted. Upgrade at {exc.upgrade_url}")

License

Apache-2.0 — see LICENSE.

Project details


Download files

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

Source Distribution

crp_comply_sdk-4.5.0.tar.gz (33.0 kB view details)

Uploaded Source

Built Distribution

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

crp_comply_sdk-4.5.0-py3-none-any.whl (25.9 kB view details)

Uploaded Python 3

File details

Details for the file crp_comply_sdk-4.5.0.tar.gz.

File metadata

  • Download URL: crp_comply_sdk-4.5.0.tar.gz
  • Upload date:
  • Size: 33.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.2

File hashes

Hashes for crp_comply_sdk-4.5.0.tar.gz
Algorithm Hash digest
SHA256 10eabae62d16b21fc7333665af9bb9fd6a7037ee10351ed10b95ab5f5790761d
MD5 f544dd6a2e8325cc958a7c3699e2d119
BLAKE2b-256 24a8285a858f3f9b3b9844c0dde858bb5b50aa8e302769de1be97c00cae60e0f

See more details on using hashes here.

File details

Details for the file crp_comply_sdk-4.5.0-py3-none-any.whl.

File metadata

  • Download URL: crp_comply_sdk-4.5.0-py3-none-any.whl
  • Upload date:
  • Size: 25.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.2

File hashes

Hashes for crp_comply_sdk-4.5.0-py3-none-any.whl
Algorithm Hash digest
SHA256 ec536c02debfe1bf3d49d1933310010cff76c656d4d812e22344c860d017ec4a
MD5 ceb128bf9f41a7739fce0a2dc733b626
BLAKE2b-256 cdcf5d72bc1351ffc6fc2c965468e726c5641b12435ac9b5081f604c6e295738

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page