mcp env gets ignored

[michabbb]

Describe the bug

same mcp works with every other cli tool so the mcp iteself is NOT the problem 😏

{
  "mcpServers": {
    "mysql_client": {
      "type": "local",
      "command": "mysql_mcp_server",
      "tools": [
        "*"
      ],
      "args": [],
      "env": {
        "MYSQL_HOST": "xxxxx",
        "MYSQL_PORT": "xxx",
        "MYSQL_USER": "xxxx",
        "MYSQL_PASSWORD": "xxxxxxxx",
        "MYSQL_DATABASE": "xxxxx"
      }
    }
  }
}

2025-09-25T22:20:49.336Z [LOG] [mcp server mysql_client stderr] 2025-09-26 00:20:49,335 - mysql_mcp_server - ERROR - Missing required database configuration. Please check environ
         │ ment variables:
     8   │ 2025-09-26 00:20:49,335 - mysql_mcp_server - ERROR - MYSQL_USER, MYSQL_PASSWORD, and MYSQL_DATABASE are required
     9   │ Traceback (most recent call last):
    10   │   File "/home/micha/.local/bin/mysql_mcp_server", line 8, in <module>
    11   │     sys.exit(main())

MCP: https://github.com/designcomputer/mysql_mcp_server

Affected version

0.0.327

Steps to reproduce the behavior

setup the mysql_mcp_server with correct env

Expected behavior

env variables should be used

Additional context

No response

[JakubPavlicek]

+1 I'm having the same problem when setting up the Oracle MCP Server: https://github.com/danielmeppiel/oracle-mcp-server
The config shown below works in every other CLI tool.

my config:

{
  "mcpServers": {
    "oracle": {
      "tools": [
        "*"
      ],
      "type": "local",
      "command": "uv",
      "args": [
        "--directory",
        "/Users/user/Downloads/oracle-mcp-server",
        "run",
        "main.py"
      ],
      "env": {
        "ORACLE_CONNECTION_STRING": "xxx",
        "TARGET_SCHEMA": "xxx",
        "CACHE_DIR": ".cache",
        "READ_ONLY_MODE": "1"
      }
    }
  }
}

[dsyme]

Same problem here

{
  "mcpServers": {
    "time": {
      "tools": [
        "printEnv"
      ],
      "type": "local",
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "-e",
        "LOCAL_TIMEZONE",
        "mcp/everything"
      ],
      "env": {
        "LOCAL_TIMEZONE": "${LOCAL_TIMEZONE}"
      }
    }
  }
}

Then

call the printEnv tool and show its result     

Gives

 ● Here are the current environment variables:

     * NODE_VERSION: 22.14.0
     * HOSTNAME: a06dccd65b65
     * YARN_VERSION: 1.22.22
     * SHLVL: 1
     * HOME: /root
     * PATH: /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
     * PWD: /app
     * NODE_ENV: production

Note that LOCAL_TIMEZONE is not there

[danielmeppiel]

When I set a MCP server, it is not getting loaded and there are no signs of what may be going on in the logs:

mcp-config.json

{
  "mcpServers": {
    "playwright-mcp": {
      "type": "local",
      "tools": [
        "*"
      ],
      "id": "41b79849-7e6c-4fc7-82c0-5a611ea21523",
      "command": "npx",
      "args": [
        "-y",
        "@playwright/mcp"
      ]
    }
  }
}

Copilot logs do not show any tools from the playwright MCP server being loaded:

2025-09-26T19:58:49.199Z [DEBUG] Tools: 
2025-09-26T19:58:49.199Z [DEBUG] [
  {
    "type": "function",
    "function": {
      "name": "bash",
      "description": "Runs a Bash command in an interactive Bash session.\n                * When invoking this tool, the contents of the \"command\" parameter does NOT need to be XML-escaped.\n* You don't have access to the internet via this tool.\n* You can run Python, Node.js and Go code with the `python`, `node` and `go` commands.\n* Each sessionId identifies a persistent Bash session. State is saved across command calls and discussions with the user.\n* `timeout` parameter must be greater than the default timeout of 30 seconds and less than 600 seconds}. Give long-running commands enough time to complete.\n* If the command does not complete within \"timeout\" seconds, the tool will return a status indicating that it is still running asynchronously. You can then use `read_bash` or `stop_bash`.\n* You can install Linux, Python, JavaScript and Go packages with the `apt`, `pip`, `npm` and `go` commands.",
      "parameters": {
        "type": "object",
        "properties": {
          "command": {
            "type": "string",
            "description": "The Bash command and arguments to run."
          },
          "description": {
            "type": "string",
            "description": "A short human-readable description of what the command does, limited to 100 characters, for example \"List files in the current directory\", \"Install dependencies with npm\" or \"Run RSpec tests\"."
          },
          "timeout": {
            "type": "integer",
            "description": "(Optional) Maximum time in seconds to wait for the command to complete when \"async\" is false. Default is 30 seconds if not provided."
          },
          "sessionId": {
            "type": "string",
            "description": "Indicates which Bash session to run the command in. Multiple sessions may be used to run different commands at the same time."
          },
          "async": {
            "type": "boolean",
            "description": "If true, runs the command asynchronously. You can send input to the command using the `write_bash` tool and read its output using the `read_bash` tool."
          }
        },
        "required": [
          "command",
          "description",
          "sessionId",
          "async"
        ]
      }
    }
  },
  {
    "type": "function",
    "function": {
      "name": "write_bash",
      "description": "Sends input to the specified command or Bash session.\n                 * This tool can be used to send input to a running Bash command or an interactive console app.\n                 * Bash commands are run in an interactive Bash session with a TTY device and Bash command processor.\n                 * sessionId (required) must match the sessionId used to invoke the async bash command.\n                 * You can send text, {up}, {down}, {left}, {right}, {enter}, and {backspace} as input.\n                 * Some applications present a list of options to select from. The selection is often denoted using ❯, >, or different formatting.\n                 * When presented with a list of items, make a selection by sending arrow keys like {up} or {down} to move the selection to your chosen item and then {enter} to select it.\n                 * The response will contain any output read after \"delay\" seconds. Delay should be appropriate for the task and never less than 10 seconds.",
      "parameters": {
        "type": "object",
        "properties": {
          "sessionId": {
            "type": "string",
            "description": "Indicates which Bash session to run the command in. Multiple sessions may be used to run different commands at the same time."
          },
          "input": {
            "type": "string",
            "description": "The input to send to the command or session."
          },
          "delay": {
            "type": "integer",
            "description": "(Optional) The amount of time in seconds to wait before reading the output that resulted from the input."
          }
        },
        "required": [
          "sessionId",
          "input"
        ]
      }
    }
  },
  {
    "type": "function",
    "function": {
      "name": "read_bash",
      "description": "Reads output from a Bash command.\n                 * Reads the output of a command running in an \"async\" Bash session.\n                 * The sessionId must be the same one used to invoke the bash command.\n                 * You can call this tool multiple times to read output produced since the last call.\n                 * Each request has a cost, so provide a reasonable \"delay\" parameter value for the task, to minimize the need for repeated reads that return no output.\n                 * If a read request generates no output, consider using exponential backoff in choosing the delay between reads of the same command.\n                 * Though `write_bash` accepts ANSI control codes, this tool does not include them in the output.",
      "parameters": {
        "type": "object",
        "properties": {
          "sessionId": {
            "type": "string",
            "description": "The ID of the shell session used to invoke the Bash command."
          },
          "delay": {
            "type": "integer",
            "description": "(Optional) The amount of time in seconds to wait before reading the output."
          }
        },
        "required": [
          "sessionId",
          "delay"
        ]
      }
    }
  },
  {
    "type": "function",
    "function": {
      "name": "stop_bash",
      "description": "Stops a running Bash command.\n                 * Stops a running Bash command by terminating the entire Bash session and process.\n                 * This tool can be used to stop commands that have not exited on their own.\n                 * Any environment variables defined will have to be redefined after using this tool if the same session ID is used to run a new command.\n                 * The sessionId must match the sessionId used to invoke the bash command.",
      "parameters": {
        "type": "object",
        "properties": {
          "sessionId": {
            "type": "string",
            "description": "The ID of the Bash session used to invoke the bash command."
          }
        },
        "required": [
          "sessionId"
        ]
      }
    }
  },
  {
    "type": "function",
    "function": {
      "name": "str_replace_editor",
      "description": "Editing tool for viewing, creating and editing files\n     * State is persistent across command calls and discussions with the user\n     * If `path` is a file, `view` displays the result of applying `cat -n`. If `path` is a directory, `view` lists non-hidden files and directories up to 2 levels deep\n     * The `create` command cannot be used if the specified `path` already exists, or if parent directories don't exist\n     * If a `command` generates a long output, output will be truncated and marked with `<file too long...`\n     * Path *must* be absolute\n\n     Notes for using the `str_replace` command:\n     * The `old_str` parameter should match EXACTLY one or more consecutive lines from the original file\n     * If the `old_str` parameter is not unique in the file, replacement will not be performed. Make sure to include enough context in `old_str` to make it unique\n     * The `new_str` parameter should contain the edited lines that should replace the `old_str`",
      "parameters": {
        "type": "object",
        "properties": {
          "command": {
            "type": "string",
            "enum": [
              "view",
              "create",
              "str_replace",
              "insert"
            ],
            "description": "The commands to run. Allowed options are: `create`, `view`, `str_replace`, `insert`."
          },
          "file_text": {
            "description": "Required parameter of `create` command; the content of the file to be created.",
            "type": "string"
          },
          "insert_line": {
            "description": "Required parameter of `insert` command. The `new_str` will be inserted AFTER the line `insert_line` of `path`.",
            "type": "integer"
          },
          "new_str": {
            "description": "Required parameter of `str_replace` command containing the new string. Required parameter of `insert` command containing the string to insert.",
            "type": "string"
          },
          "old_str": {
            "description": "Required parameter of `str_replace` command containing the string in `path` to replace. Leading and ending whitespaces from file content should be preserved!",
            "type": "string"
          },
          "path": {
            "description": "Absolute path to file or directory.",
            "type": "string"
          },
          "view_range": {
            "description": "Optional parameter of `view` command when `path` points to a file. If none is given, the full file is shown. If provided, the file will be shown in the indicated line number range, e.g. [11, 12] will show lines 11 and 12. Indexing at 1 to start. Setting `[start_line, -1]` shows all lines from `start_line` to the end of the file.",
            "items": {
              "type": "integer"
            },
            "type": "array"
          }
        },
        "required": [
          "command",
          "path"
        ]
      }
    },
    "copilot_cache_control": {
      "type": "ephemeral"
    }
  }
]
2025-09-26T19:58:49.199Z [END-GROUP] 

[michabbb]

Okay, it's obviously a general problem, so please, people, stop posting your configs 😏 thanks.

[pelikhan]

this is a show-stopper to use mcps that require secrets.

[michabbb]

this is a show-stopper to use mcps that require secrets.

Really? I thought we created that issue here just for fun but glad you made that clear 👍

[RyanHecht]

Hey folks! We're looking into this bug, thank you all for the reports

[williammartin]

In most cases this is almost certainly in most cases a result of the env section being a pass through/mapping of environment variables provided to copilot, rather than using the values in the config directly, as described in https://docs.github.com/copilot/how-tos/agents/copilot-coding-agent/extending-copilot-coding-agent-with-mcp

For example:

{
  "mcpServers": {
    "github-mcp-server-stdio": {
      "type": "local",
      "command": "docker",
      "tools": [
        "*"
      ],
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "GITHUB_PERSONAL_ACCESS_TOKEN=$GITHUB_MCP_TOKEN",
        "ghcr.io/github/github-mcp-server"
      ],
      "env": {
        "GITHUB_MCP_TOKEN": "GITHUB_MCP_TOKEN"
      }
    }
  }
}

The env section is not indicating "set GITHUB_MCP_TOKEN to the value" but "pass through the value of GITHUB_MCP_TOKEN in copilot's env to the GITHUB_MCP_TOKEN in the environment of the executed command".

[pelikhan]

(GITHUB_* are reserved secret names, this example won't work).

The docs indicate that there is some name mangling with a COPILOT_MCP_ prefix. Is this the case or can we pass any secret identifier?

The name of a GitHub Actions secret you have configured, beginning with COPILOT_MCP_ preceded by a $