---
title: ModelsLab Developer Resources - OpenAPI Specification, MCP Server, API Scopes
description: Every machine-readable ModelsLab developer resource in one place: the Generation and Agent Control Plane OpenAPI 3.1 specifications, MCP server manifest, agent authentication guide, API scope catalogue, rate-limit conventions and deprecation policy.
url: https://modelslab.com/developers/resources
canonical: https://modelslab.com/developers/resources
type: website
component: Business/DeveloperResources
generated_at: 2026-08-24T07:53:43.039156Z
---

ModelsLab developer resources
---

Every machine-readable contract ModelsLab publishes, at a stable URL. If you are an agent, start with [llms.txt](https://modelslab.com/llms.txt) or the [API catalog](https://modelslab.com/.well-known/api-catalog); both link to everything below.

ModelsLab Generation API — OpenAPI 3.1 Specification
---

OpenAPI 3.1 · application/json

The complete contract for image, video, audio, 3D, training and LLM generation across /api/v6, /api/v7 and /api/v8. Typed request and response schemas on all 133 operations, each with a unique operationId, a description and typed parameters.

<https://modelslab.com/openapi.json>

ModelsLab Agent Control Plane — OpenAPI 3.1 Specification
---

OpenAPI 3.1 · application/json

Account, billing, wallet, subscription, team, API-key and usage operations at /api/agents/v1. Every operation declares the token scope it requires as x-required-scope, so an agent can compute its minimum privilege set before authenticating.

<https://modelslab.com/agents-openapi.json>

ModelsLab MCP Server Manifest
---

MCP manifest · application/json

Three Model Context Protocol servers over Streamable HTTP: generation (v7 and v8) and the agent control plane. Each exposes tools plus reference resources — the error model, rate-limit policy, scope catalogue and site index. The SEP-1649 server card is at /.well-known/mcp/server-card.json.

<https://modelslab.com/.well-known/mcp.json>

ModelsLab API Catalog (RFC 9727 linkset)
---

RFC 9727 · application/linkset+json

One machine-readable entry point that links every API surface to its specification, documentation, status page, terms, privacy policy and deprecation policy.

<https://modelslab.com/.well-known/api-catalog>

ModelsLab Agent Authentication Guide
---

auth.md · text/markdown

The registration runbook for autonomous agents: discovery, signup, the email claim ceremony, exchanging credentials for a scoped control-plane token, the error table and revocation.

<https://modelslab.com/auth.md>

ModelsLab OAuth Metadata
---

RFC 8414 and RFC 9728 · application/json

Authorization server and protected resource metadata, including the full list of supported scopes and the endpoint that describes them.

<https://modelslab.com/.well-known/oauth-authorization-server>

ModelsLab Site Index for LLMs
---

llms.txt · text/plain

The agent-facing index of the whole platform: search entry points, markdown access, all five integration paths, the developer-resource index, scopes, rate limits and the error model. The expanded form is at /llms-full.txt.

<https://modelslab.com/llms.txt>

ModelsLab Model Catalog Feed
---

Catalog feed · application/json

Every publicly listed model with its identifier, category, price, billing unit and plan coverage. Paginated and filterable by category.

<https://modelslab.com/models-feed>

ModelsLab CLI
---

CLI · Homebrew, Scoop, install.sh

The official command-line interface. Install with brew install modelslab/tap/modelslab, scoop, or curl -fsSL https://modelslab.sh/install.sh | sh. Covers authentication, billing, wallet, subscriptions, model discovery and generation.

<https://modelslab.sh>

ModelsLab Agent Skills
---

Agent skills · npx skills

Skill files that install straight into a coding agent. Install with npx skills add modelslab/skills --all.

<https://github.com/ModelsLab/skills>

ModelsLab API scopes
---

Control-plane tokens support least-privilege scopes. Send a `scopes` array on `POST /api/agents/v1/auth/login`; omit it and the token is issued with `*`, which is the historical behaviour. A call outside your scopes returns `403` with `error.code = insufficient_scope`, naming the required scope in the body and in the `WWW-Authenticate` header. Self-revocation is never scope-gated, and switching account context can never widen scopes.

| Scope | Grants |
|---|---|
| \* | Full account access. Equivalent to every scope below; the default when no scopes are requested. |
| profile:read | Read the account profile, email and preferences. |
| profile:write | Update the account profile, password, socials and preferences. |
| tokens:read | List the access tokens issued for this account. |
| tokens:write | Revoke access tokens, log out, and switch account context. |
| api-keys:read | List and inspect generation API keys. |
| api-keys:write | Create, rename and delete generation API keys. |
| usage:read | Read usage summaries, per-product usage and request history. |
| models:read | Browse the model catalogue, filters, tags and providers. |
| files:write | Upload files and base64 payloads for use as generation inputs. |
| billing:read | Read billing info, invoices, payment methods and the Stripe publishable key. |
| billing:write | Add and remove payment methods, update billing info and create payment links. |
| wallet:read | Read the wallet balance, coupon validity and payment status. |
| wallet:write | Fund the wallet, withdraw, redeem coupons and change auto-funding. |
| subscriptions:read | Read plans, current subscriptions and subscription status. |
| subscriptions:write | Buy, change, pause, resume and repair subscriptions. |
| teams:read | List teams and team members. |
| teams:write | Create and modify teams, invite members and accept invitations. |

ModelsLab rate limits and API lifecycle
---

Every API response carries `RateLimit-Limit`, `RateLimit-Remaining`, `RateLimit-Reset` and the structured `RateLimit-Policy` and `RateLimit` fields, with `Retry-After` when the limit is exhausted. Throttle against `RateLimit-Remaining` rather than retrying blind. Versioned responses also carry `ModelsLab-Api-Version` and `ModelsLab-Api-Lifecycle`; a deprecated version adds `Deprecation` (RFC 9745) and `Sunset` (RFC 8594) with a minimum of 90 days between them.

Full policy: [ModelsLab API versioning and deprecation policy](/api-deprecation-policy).

---

*This markdown version is optimized for AI agents and LLMs.*

**Links:**
- [Website](https://modelslab.com)
- [API Documentation](https://docs.modelslab.com)
- [Blog](https://modelslab.com/blog)

---
*Generated by ModelsLab - 2026-08-24*