Skip to main content

Correctover — LLM Reliability Engineering platform. Protocol-level contract validation, verified failover (DeepSeek/OpenAI/Anthropic), MCP security, CCS v1.0 standard, 84.1% self-heal rate across 20K+ trajectories.

Project description

Correctover SDK

Failure is not fatal. Protocol-level contract validation with automatic verified failover for LLM APIs — now with MCP security, CCS v1.0 standard, and runtime guardrails.

PyPI version Python 3.12+ License: Proprietary

Install

pip install correctover

Quick Start

import correctover

# Automatic failover with verified contract validation
result = correctover.run("Hello!")
print(result)

What It Does

Correctover is the LLM Reliability Engineering (LLMRE) platform for AI agents:

Capability Description
Contract Validation Protocol-level output verification — schema, semantic, safety
Verified Failover Automatic retry to DeepSeek/OpenAI/Anthropic on failure
Self-Healing 84.1% success rate across 20K+ decision trajectories
MCP Security 14 OWASP AISVS 1.0 checks for Model Context Protocol configs
Runtime Guard 22µs P50 RCE/SSRF/credential leak interception
CCS v1.0 Standard First formal conformance standard for agentic runtimes (DOI: 10.5281/zenodo.21234580)

CLI Tools

correctover-shield    # Runtime guard daemon
correctover-ccs       # CCS v1.0 conformance scanner
correctover-mcs       # MCP security scanner
correctover-hunt      # Vulnerability hunter

Key Metrics

Metric Value
Self-heal success rate 84.1%
Failover latency 0.4–0.7ms
Runtime guard P50 22µs
Test trajectories 20,071
CVEs filed 6 (AutoGen ×2, CrewAI, Semantic Kernel, MS AGT)
Fault types 215

Supported Providers

  • DeepSeek (primary failover target)
  • OpenAI (GPT-4o, GPT-4-turbo)
  • Anthropic (Claude 3.5 Sonnet)

Free Tier

50 calls/day — no credit card required.

Unlock unlimited: correctover.com/checkout

export CORRECTOVER_LICENSE_KEY=your-key-here

Related Correctover Tools

Tool Install Description
Security Scanner npx correctover-scan MCP config security audit (14 checks)
Self-Healing Test pip install correctover-test Agent self-healing test suite
Vulnerability Scan pip install correctover-security-audit 215 fault type scanner
Compliance Check pip install correctover-compliance-check OAuth 2.1 + CCS v1.0
Runtime Guard pip install correctover-runtime-guard 22µs RCE/SSRF interception
MCP Server npm install correctover-mcp-server 6-dimension validation

Links

License

Proprietary Commercial License. See LICENSE file.

Project details


Download files

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

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

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

correctover-2.3.2-py3-none-any.whl (248.5 kB view details)

Uploaded Python 3

File details

Details for the file correctover-2.3.2-py3-none-any.whl.

File metadata

  • Download URL: correctover-2.3.2-py3-none-any.whl
  • Upload date:
  • Size: 248.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.12

File hashes

Hashes for correctover-2.3.2-py3-none-any.whl
Algorithm Hash digest
SHA256 a3024a340d3b7de3271e3c35cd635c49b64871b40d3dbf66406afdb752f1244a
MD5 4dd97bb1293c4def381ac008bba9587c
BLAKE2b-256 a3f8f026e1a96e37ce89f4e6fcda1f773f2915d5712fbabf2cb9966864855275

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