canonical: https://jentic.com/apis/mandapi.com/mandapi # MandAPI The MandAPI API offers an OpenAI-compatible chat completion endpoint that reaches GPT, Claude, Gemini, DeepSeek, and other models from a single API key. A models endpoint lists what the account can currently access, and the completion request names the model to use, so an agent can switch between providers by changing one field rather than integrating each vendor separately. Model availability and supported parameters depend on the chosen model and account. ## For AI agents Generate chat completions from GPT, Claude, Gemini, DeepSeek, and other models through one OpenAI-compatible endpoint and a single key, and list the models the account can use. ## Scope Does not fine-tune models, generate images, or store conversation history. Use for listing models and generating chat completions only. ## Capabilities - Generate a chat completion from a conversation of messages - Reach GPT, Claude, Gemini, DeepSeek, and other models with one API key - Select the model for a request by naming it in the completion call - List the models the account can currently access ## Use cases ### One key for many models An AI agent that needs to try different providers calls a single OpenAI-compatible endpoint and names the model per request. The MandAPI API reaches GPT, Claude, Gemini, DeepSeek, and other models from one key, so the agent can compare or fail over between providers without wiring each vendor's SDK. Through Jentic the completion operation is discovered by intent and the key is injected at call time. Example prompt: Send a chat completion request naming a specific model and read back the generated reply ### Model discovery before a call Because availability depends on the account, an agent can list the models it may use before sending a request. The MandAPI API returns the current model list from its models endpoint, so a workflow can pick an available model dynamically instead of hard-coding a name that may not be enabled. Example prompt: List the available models and choose one before issuing a chat completion request ### Drop-in OpenAI-compatible completions Teams already using OpenAI-shaped requests can point an agent at MandAPI with minimal change. The MandAPI API accepts the same chat completion request and response shape, so existing message-based prompts work while the model field selects which provider answers. Example prompt: Reuse an existing OpenAI-shaped messages payload against MandAPI and verify the completion returns ## Key resources - **Chat** — Create a chat completion from a conversation of messages, naming the model to use - **Models** — List the models the account can currently access ## AI readiness This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use. - **Score:** 67 / 100 - **Maturity:** AI-Aware - **Dimensions:** - Foundational Compliance: 100 / 100 - Developer Experience & Jentic Compatibility: 71 / 100 - AI-Readiness & Agent Experience: 43 / 100 - Agent Usability: 94 / 100 - Security: 60 / 100 - AI Discoverability: 88 / 100 - **View full report:** https://jentic.com/apis/mandapi.com/mandapi/scorecard - **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/ - **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview ### Score it yourself Every API in the directory is allowlisted, so you can re-score it with no key required. - **Score your own API:** https://jentic.com/scorecard.md - **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md ```sh npx @jentic/api-scorecard-cli score ``` ## Why Jentic - **Setup:** Wiring the MandAPI API by hand means handling the bearer key, matching the OpenAI-compatible request shape, and checking per-account model availability yourself. Through Jentic you install once, import it from the API Directory, store the key once, and your agent generates completions through it. - **Permission scoping:** MandAPI carries the model choice inside the request body, so rules bound which operations your agent may call, such as generating completions and listing models, rather than which model. You choose the operation set, so nothing beyond those two operations is reachable unless you add it. - **Credential handling:** Your MandAPI key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context. - **Discovery method:** Agents search Jentic by intent such as 'generate a chat completion' or 'list available models', and Jentic returns the matching MandAPI operation with its input schema so the agent calls the right endpoint without browsing the reference docs. ## Related APIs - **OpenAI** — OpenAI serves its own models directly, where MandAPI fronts several providers behind one key. - **Mistral AI** — Mistral AI serves its own open-weight and hosted models through a dedicated API. - **Groq** — Groq runs supported models at very low latency on its own hardware. ## FAQ ### What authentication does the MandAPI API use? The MandAPI API authenticates with a bearer token that carries your API key, per its OpenAPI spec, matching the OpenAI-style Authorization header. Through Jentic the key is stored once and injected at call time, so it never enters the agent's prompt, logs, or context. ### Which models can I reach with the MandAPI API? The completion endpoint reaches GPT, Claude, Gemini, DeepSeek, and other models from a single key, and the model is chosen per request. Because availability depends on your account, list the models endpoint first to see which ones you can currently use. ### What are the rate limits for the MandAPI API? The OpenAPI spec does not define numeric rate limits, and the limits that apply can vary by model and account. Have your agent back off when it receives an HTTP 429 response and check your account dashboard for the current limits. ### Is there a MandAPI MCP server? You don't need an MCP server to give your agent the MandAPI API. Jentic connects it directly from the API Directory: import it, store your key once, and your agent generates completions through it. Operations are discovered on demand, so no extra tool definitions sit in the agent's context. ### Can I limit what my agent is allowed to do with the MandAPI API? Yes. MandAPI carries the model choice inside the request body, so rules bound which operations your agent may call, such as generating completions and listing models, rather than which model. You choose the operation set, and every call it makes is logged. ### How do I generate a chat completion through Jentic? Search Jentic for an intent such as 'generate a chat completion' and it returns the matching completion operation with its input schema, so your agent passes the messages and model name and reads back the reply. To run it on your own infrastructure, install Jentic One from its GitHub repo.