bug: Error: Required API key PERPLEXITY_API_KEY for provider 'perplexity' is not set in environment, session, or .env file. #722

New issue

Closed

opened 2025-10-14 16:10:52 -06:00 by navan · 0 comments

navan commented

2025-10-14 16:10:52 -06:00

Owner

Originally created by @celgost on 5/16/2025

Description

This is not very clear in documentation how to actually use another API in Cursor.

I am eager to make a small PR to update documentation if you tell me how to use another API within Cursor MCP implementation.

I assumed I needed to:

Provide my OpenAI key in mcp.json.
Use provider openai and the modelId in config.

I also removed any mention of the research flag in the auto-generated cursor rules. But it doesn't matter.

See files below.

Steps to Reproduce

Just ask in the Agent chat of Cursor to go to the next task and update task.
Call is made automatically by the Agent in chat, like seen in screenshot.
(I also tried to change 4.1-mini to 4-1-mini like in models list)

Expected Behavior

When updating tasks, the MCP tool checks the API key in mcp.json and the configured models in .taskmasterconfig and then use the providers and models defined there.

Actual Behavior

Error: Required API key PERPLEXITY_API_KEY for provider 'perplexity' is not set in environment, session, or .env file.

Environment

Task Master version: latest
Node.js version: 22.14.0
Operating system: Ubuntu 24.04 LTS
IDE (if applicable): Cursor
Model Used in Cursor: Gemini 2.5 Pro Exp 03 25

Additional Context

mcp.json:

"mcpServers": {
    "taskmaster-ai": {
      "command": "npx",
      "args": ["-y", "--package=task-master-ai", "task-master-ai"],
      "env": {
        "OPENAI_API_KEY": "sk-proj-myapikey"
      }
    }
}

.taskmasterconfig

{
  "models": {
    "main": {
      "provider": "openai",
      "modelId": "gpt-4-1-mini",
      "maxTokens": 120000,
      "temperature": 0.2
    },
    "research": {
      "provider": "openai",
      "modelId": "gpt-4-1-mini",
      "maxTokens": 8700,
      "temperature": 0.1
    },
    "fallback": {
      "provider": "openai",
      "modelId": "gpt-4-1-mini",
      "maxTokens": 120000,
      "temperature": 0.1
    }
  },
  "global": {
    "logLevel": "info",
    "debug": false,
    "defaultSubtasks": 5,
    "defaultPriority": "medium",
    "projectName": "Project"
  }
}

*Originally created by @celgost on 5/16/2025* ### Description This is not very clear in documentation how to actually use another API in Cursor. I am eager to make a small PR to update documentation if you tell me how to use another API within Cursor MCP implementation. I assumed I needed to: - Provide my OpenAI key in `mcp.json`. - Use provider `openai` and the `modelId` in config. I also removed any mention of the research flag in the auto-generated cursor rules. But it doesn't matter. See files below. ### Steps to Reproduce 1. Just ask in the Agent chat of Cursor to go to the next task and update task. 2. Call is made automatically by the Agent in chat, like seen in screenshot. (I also tried to change 4.1-mini to 4-1-mini like in models list) ### Expected Behavior When updating tasks, the MCP tool checks the API key in `mcp.json` and the configured models in `.taskmasterconfig` and then use the providers and models defined there. ### Actual Behavior ![Image](https://github.com/user-attachments/assets/064a7906-9ef2-4159-90a2-7181861e5a90) ![Image](https://github.com/user-attachments/assets/d2522f98-8550-438a-a00c-c59eadc83ff3) Error: Required API key PERPLEXITY_API_KEY for provider 'perplexity' is not set in environment, session, or .env file. ### Environment - Task Master version: latest - Node.js version: 22.14.0 - Operating system: Ubuntu 24.04 LTS - IDE (if applicable): Cursor - Model Used in Cursor: Gemini 2.5 Pro Exp 03 25 ### Additional Context mcp.json: ```json "mcpServers": { "taskmaster-ai": { "command": "npx", "args": ["-y", "--package=task-master-ai", "task-master-ai"], "env": { "OPENAI_API_KEY": "sk-proj-myapikey" } } } ``` .taskmasterconfig ``` { "models": { "main": { "provider": "openai", "modelId": "gpt-4-1-mini", "maxTokens": 120000, "temperature": 0.2 }, "research": { "provider": "openai", "modelId": "gpt-4-1-mini", "maxTokens": 8700, "temperature": 0.1 }, "fallback": { "provider": "openai", "modelId": "gpt-4-1-mini", "maxTokens": 120000, "temperature": 0.1 } }, "global": { "logLevel": "info", "debug": false, "defaultSubtasks": 5, "defaultPriority": "medium", "projectName": "Project" } } ```