Repository-based API discovery
Scan repositories to identify APIs and endpoints, build a catalog, and generate OpenAPI 3.1 specifications from code without requiring an existing OpenAPI file.
Elva
Freemium
Elva is an API management platform that discovers APIs from source repositories, evaluates their design and AI readiness, and governs how they are published to developers and AI agents. It generates contracts, documentation, MCP servers, and related delivery artifacts without requiring an existing OpenAPI specification.
Elvaとは？
Elva is an API management platform for teams that need to discover, assess, govern, test, and publish APIs for different audiences. It scans connected repositories to build an API catalog and can generate OpenAPI specifications from code, so teams can begin without maintaining a separate specification first.
The platform scores APIs and endpoints across areas such as design, developer experience, AI readiness, security, and performance. It can identify issues including missing descriptions, untyped response fields, and undocumented authentication, then provide AI-assisted fixes where available. API testing includes manual testing, with automated scheduled and CI testing and chaos suites on higher plans.
Elva uses audience-specific API contracts to define which endpoints and fields are exposed and where the resulting interface is published. A contract can produce specifications, documentation, MCP servers, Postman collections, and TypeScript SDKs depending on the plan and deployment configuration. Elva also compares source changes against contracts and supports breaking-change policies, notifications, governance rules, and approval workflows on higher tiers.
For AI-agent access, Elva generates hosted or self-hosted MCP servers from governed contracts. Its MCP gateway checks identity and tool scope, applies rate and quota controls, can redact fields according to the contract, and records calls. Teams can test servers in the MCP Playground and use agent feedback and operational insights to improve the catalog and exposed tools.
Elvaでできること
API scoring and AI-assisted fixes
Evaluate endpoints and collections across design, developer experience, AI readiness, security, and performance; identify documentation, schema, and authentication gaps and apply supported AI fixes.
Audience-specific API contracts
Select audiences, endpoints, and fields for internal, partner, public, or AI-agent access. Contracts can define visibility, approval requirements, and field exclusions such as PII or internal data.
Change governance and publishing
Compare commits against contract promises, surface schema and authentication changes, and support breaking-change policies, stakeholder notifications, rulesets, webhooks, and contract deployment targets on applicable plans.
Governed MCP servers
Generate curated MCP tools from API contracts rather than exposing an entire API. Elva’s gateway supports identity checks, per-tool scopes, rate and quota controls, field redaction, and call logging.
Testing and observability
Use the MCP Playground for interactive testing, collect agent feedback, and inspect server status and call activity. Business and Enterprise plans add automated API testing, chaos suites, and expanded MCP insights.
利用シーン
“Create a catalog from undocumented code”
An API team can connect a repository, discover endpoints, generate a baseline specification, and review scores and gaps before investing in a manually maintained catalog.
“Publish a controlled partner API”
A team can define a partner contract that selects approved endpoints and fields, excludes internal or sensitive data, requires approval, and publishes the resulting specification and documentation.
“Expose selected APIs to AI agents”
A platform or security team can generate an MCP server from a contract, expose only suitable tools, assign per-agent or per-key scopes, and inspect calls through the gateway logs.
“Prevent breaking changes”
API owners can compare source changes with published contracts, review schema or authentication diffs, notify stakeholders, and use governance policies to block or warn on changes that affect consumers.
“Improve agent-facing tools”
Developers can exercise an MCP server in the Playground, inspect the calls made by an agent, and use feedback about unclear descriptions, timeouts, or parameter complexity to improve the catalog.
よくある質問
Does Elva require an OpenAPI specification?
No. The site states that Elva can read repositories, discover APIs from code, and generate an OpenAPI specification. An existing OpenAPI spec is not required to start.
What can an Elva contract control?
A contract can specify the audience, selected endpoints, exposed fields, and publishing destinations. The documented audiences include internal services, partners, public APIs, and AI agents using MCP. Contracts can also support approval, breaking-change, notification, and governance policies depending on the plan.
How are MCP servers secured?
Elva places MCP tool calls behind an authorization gateway. The documented checks include identity, tool-level scope, field redaction, rate and quota controls, and an audit record containing call details such as the agent, tool, latency, and outcome. Hosted and self-hosted deployment options are described, with self-hosted deployment available on Enterprise.
Can teams test an MCP server before using it in production?
Yes. The MCP Playground lets teams interact with their server through an agent and inspect the tool calls and responses. The site states that Playground calls do not count against the MCP gateway quota.
Is there a free version of Elva?
Yes. The pricing page lists a Free plan and paid Startup, Business, and Enterprise plans. It also states that plans have unlimited members and no per-seat pricing; exact repository, server, call, governance, and testing limits vary by plan.
クイック情報
- Category
- API management and developer tools
- Primary users
- API, platform, developer experience, and security teams
- Core workflow
- Repository scan → API catalog and scoring → audience contract → governed publishing
- MCP deployment
- Hosted MCP servers are available; self-hosted deployment is listed for Enterprise
- API specification
- OpenAPI generation from code; an existing OpenAPI specification is not required
- Pricing model
- Free tier plus Startup, Business, and custom Enterprise plans; unlimited members and no per-seat pricing
Elva のトラフィック分析
トラフィックデータは参考情報としてご利用ください。
- ドメイン評価
- 8
Elvaの代替品
Cadenya
www.cadenya.com
Cadenyaは、既存のAPIを活用してAIエージェントを構築、テスト、改善できる開発者向けのエージェント実行環境です。ツール、モデル、目標、webhook、フィードバック、埋め込み可能なウィジェットを一つのプラットフォームで連携できます。
Relaticle
relaticle.com
Relaticleは、共有レコード、パイプラインとタスクの管理、組み込みのAIチャット、39種類のツールを備えたMCPサーバーを統合した、チーム向けのオープンソースCRMです。ホスト型ワークスペースを利用するか、AGPL-3.0のアプリケーションをセルフホストし、MCPとREST API経由で外部AIエージェントに接続できます。
Openship
openship.io
Openship is an open-source deployment platform for building and running applications from Git repositories, local folders, or templates. Teams can use its dashboard, CLI, desktop app, or MCP interface to deploy on Openship Cloud, connected servers, or their own infrastructure.
WebDecoy
webdecoy.dev
WebDecoy は、Vercel のサイトやアプリに届く自動化トラフィックを監視し、クローラー、AI ボット、ツール、user agent のないリクエストを識別します。公開済み IP 範囲または署名付きリクエストを検証し、偽装されたよく知られた user agent にフラグを立て、対応する Vercel プランではアプリケーションコードを変更せずに活動状況を報告します。