# Cleaninglab

**Canonical:** https://apis.io/providers/cleaninglab/  
**Website:** https://cleanlab.ai  
**APIs profiled:** 1

Cleanlab provides AI safety tools to detect and remediate hallucinations, retrieval errors, and policy violations in AI agents. The platform offers guardrails, monitoring, and automated evaluation for LLM outputs, helping enterprises deploy trustworthy AI. Acquired by Handshake AI, Cleanlab serves developers with APIs, SDKs, and a web UI for real‑time error correction and compliance.

## Kin Score — 43.4 / 100 (developing)

Scored 2026-10-09 under rubric 0.23.0.

| Facet | Score |
|---|---|
| Discoverability | 69.6 |
| Contract Quality | 35.5 |
| Contract Governance | 65.5 |
| Operational Transparency | 15.8 |
| Developer Ergonomics | 42.9 |
| Access Clarity | 36.8 |

Regulatory layer — **Horizontal (data, software, accessibility, platform)**: 27.3 (matched via fallback).

## Agent readiness — 16.1 (agent-aware)

| Dimension | Value |
|---|---|
| Spec Presence | yes |
| Agentic Access | no |
| Reversibility Documented | no |
| MCP Server | documented |
| Auth Clarity | no |
| Idempotency | no |
| Error Semantics | no |
| OpenAPI Examples | no |
| Rate Limit Signal | no |
| Event Surface Described | no |
| Agent Skills | no |
| Well Known Catalog | no |
| Consent Identity | no |
| Agent Card | no |
| Dry Run Mode | no |
| Delegated Identity | no |
| Protected Resource Metadata | no |
| Dynamic Client Registration | no |
| Agentic Commerce | no |

## APIs (1)

- **Cleaninglab API** — Cleaninglab API as documented publicly: 1 operations. Contract generated from the documentation by API Evangelist (2026-10-09); not the provider's own document.

## Security (3)

- **Cleaninglab Domain Security** — TLSv1.3 · HSTS · DMARC
- **Cleaninglab Vulnerability Disclosure** — security.txt · contact published
- **Cleaninglab Trust Center** — SOC 2

## Tags

Company, AI, Safety, Monitoring, Hallucination

---

Profiled by [API Evangelist](https://apievangelist.com) and published on [APIs.io](https://apis.io/providers/cleaninglab/). Scores are computed from the provider's own public artifacts under a published rubric.