# Berrer

**Canonical:** https://apis.io/providers/berrergate-com/  
**Website:** https://berrergate.com/  
**APIs profiled:** 2

Berrer runs BerrerGate (berrergate.com), an agent-native "tool and provider intelligence" service: an AI agent describes a task in natural language and BerrerGate maps it to agent-usable tools, APIs and provider categories, returns a free, cache-only procurement preview, and sells tightly bounded "decision packets" — evidence-backed shortlists of commercially cleared open-source implementations for observability, orchestration, scraping, analytics and browser automation — plus a paid Evidence Brief and a low-cost pay-per-call AI answer, all settled in USDC on Base through x402 with no account. The same surface is published three ways from api.berrergate.com: a 25-operation OpenAPI 3.1 contract at /openapi.json, an A2A agent card at /.well-known/agent-card.json with twelve skills (listed as Verified in the Global A2A Registry), and a remote MCP endpoint at /mcp declared in an MCP server.json manifest; discovery is rounded out by an x402 manifest, an AI Registry catalog, an llms.txt and a provider-written skill.md. The company states it is experimental, publishes no terms of service, no contact address and no pricing page.

## Kin Score — 43.9 / 100 (developing)

Scored 2026-10-09 under rubric 0.23.0. Trend: flat (+2.1 from 41.8).

| Facet | Score |
|---|---|
| Discoverability | 71.7 |
| Contract Quality | 40.1 |
| Contract Governance | 18.2 |
| Operational Transparency | 31.6 |
| Developer Ergonomics | 47.6 |
| Access Clarity | 52.6 |

Regulatory layer — **Horizontal (data, software, accessibility, platform)**: 22.7 (matched via fallback).

## Agent readiness — 37.6 (agent-ready)

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

## APIs (8)

- **BerrerGate MCP Server** — Remote Model Context Protocol endpoint at https://api.berrergate.com/mcp (Streamable HTTP, POST only), declared in the provider's llms.txt, in an MCP server.json manifest at /.w...
- **BerrerGate A2A Agent** — Agent2Agent surface: an agent card served from https://api.berrergate.com/.well-known/agent-card.json (content-type application/a2a+json, version 2.9.0) declaring two interfaces...
- **Berrer A2a API** — The A2a API from Berrer — 2 operation(s) for a2a.
- **Berrer Agent API** — The Agent API from Berrer — 18 operation(s) for agent.
- **Berrer Research API** — The Research API from Berrer — 1 operation(s) for research.
- **Berrer Search API** — The Search API from Berrer — 1 operation(s) for search.
- **Berrer Skill.md API** — The Skill.md API from Berrer — 1 operation(s) for skill.md.
- **Berrer .well Known API** — The .well Known API from Berrer — 2 operation(s) for .well known.

## MCP servers (2)

- **Berrer MCP Server** — BerrerGate publishes ONE remote MCP server at https://api.berrergate.com/mcp, declared in three places the provider controls: the API host's llms.txt ("MCP: POST https://api.ber...
- **MCP server.json manifest (provider-hosted)**

## Agentic access (1)

- **Berrergate Com Agentic Access** — 25 operations · 9 acting · 1 human-in-the-loop

## Security (2)

- **Berrergate Com Authentication** — 0 schemes
- **Berrergate Com Domain Security** — TLSv1.3 · HSTS · DMARC

## Plans (1)

- **Berrergate Com Plans Pricing**

## Tags

Agents, Agentic Commerce, A2A, MCP, x402, Procurement, Tool Discovery, API Discovery, Inference, Research, Agent-Native

---

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