bug: Telescope and fzf don't use the current worspace for file picking

[HadyMash]

Describe the bug

Telescope and fzf ignore the current workspace and use the parent directory (i believe it uses the git project root). This results in an incorrect relative file path which makes the apply functionality create a new file instead of modifying the correct one.

Here is mini.pick:

And telescope/fzf:

To reproduce

Not exactly sure but i believe it uses the git repository root, so create a git repo, then create a project in a subdirectory and open neovim there. then try adding files

Expected behavior

Telescope and fzf should show relative paths and files from the current open directory and not parent directories just like mini.pick

Installation method

Using lazy.nvim:

I've tried both the default config and one i've customised:

Default config from README

{
  "yetone/avante.nvim",
  event = "VeryLazy",
  lazy = false,
  version = false, -- Set this to "*" to always pull the latest release version, or set it to false to update to the latest code changes.
  opts = {
    -- add any opts here
    -- for example
    provider = "openai",
    openai = {
      endpoint = "https://api.openai.com/v1",
      model = "gpt-4o", -- your desired model (or use gpt-4o, etc.)
      timeout = 30000, -- timeout in milliseconds
      temperature = 0, -- adjust if needed
      max_tokens = 4096,
    },
  },
  -- if you want to build from source then do `make BUILD_FROM_SOURCE=true`
  build = "make",
  -- build = "powershell -ExecutionPolicy Bypass -File Build.ps1 -BuildFromSource false" -- for windows
  dependencies = {
    "stevearc/dressing.nvim",
    "nvim-lua/plenary.nvim",
    "MunifTanjim/nui.nvim",
    --- The below dependencies are optional,
    "echasnovski/mini.pick", -- for file_selector provider mini.pick
    "nvim-telescope/telescope.nvim", -- for file_selector provider telescope
    "hrsh7th/nvim-cmp", -- autocompletion for avante commands and mentions
    "ibhagwan/fzf-lua", -- for file_selector provider fzf
    "nvim-tree/nvim-web-devicons", -- or echasnovski/mini.icons
    "zbirenbaum/copilot.lua", -- for providers='copilot'
    {
      -- support for image pasting
      "HakonHarnes/img-clip.nvim",
      event = "VeryLazy",
      opts = {
        -- recommended settings
        default = {
          embed_image_as_base64 = false,
          prompt_for_file_name = false,
          drag_and_drop = {
            insert_mode = true,
          },
          -- required for Windows users
          use_absolute_path = true,
        },
      },
    },
    {
      -- Make sure to set this up properly if you have lazy=true
      'MeanderingProgrammer/render-markdown.nvim',
      opts = {
        file_types = { "markdown", "Avante" },
      },
      ft = { "markdown", "Avante" },
    },
  },
}

Config i've customised

  -- avante
  {
    "yetone/avante.nvim",
    event = "VeryLazy",
    lazy = false,
    version = false, -- Set this to "*" to always pull the latest release version, or set it to false to update to the latest code changes.
    opts = {
      ---@alias Provider "claude" | "openai" | "azure" | "gemini" | "cohere" | "copilot" | string
      provider = "copilot", -- Recommend using Claude
      copilot = {
        model = "claude-3.5-sonnet",
        temperature = 0,
        max_tokens = 8192,
      },

      -- WARNING: Since auto-suggestions are a high-frequency operation and therefore expensive,
      -- currently designating it as `copilot` provider is dangerous because: https://github.com/yetone/avante.nvim/issues/1048
      -- Of course, you can reduce the request frequency by increasing `suggestion.debounce`.
      auto_suggestions_provider = "claude",
      claude = {
        endpoint = "https://api.anthropic.com",
        model = "claude-3-5-sonnet-20241022",
        temperature = 0,
        max_tokens = 4096,
      },
      ---Specify the special dual_boost mode
      ---1. enabled: Whether to enable dual_boost mode. Default to false.
      ---2. first_provider: The first provider to generate response. Default to "openai".
      ---3. second_provider: The second provider to generate response. Default to "claude".
      ---4. prompt: The prompt to generate response based on the two reference outputs.
      ---5. timeout: Timeout in milliseconds. Default to 60000.
      ---How it works:
      --- When dual_boost is enabled, avante will generate two responses from the first_provider and second_provider respectively. Then use the response from the first_provider as provider1_output and the response from the second_provider as provider2_output. Finally, avante will generate a response based on the prompt and the two reference outputs, with the default Provider as normal.
      ---Note: This is an experimental feature and may not work as expected.
      dual_boost = {
        enabled = false,
        first_provider = "openai",
        second_provider = "claude",
        prompt = "Based on the two reference outputs below, generate a response that incorporates elements from both but reflects your own judgment and unique perspective. Do not provide any explanation, just give the response directly. Reference Output 1: [{{provider1_output}}], Reference Output 2: [{{provider2_output}}]",
        timeout = 60000, -- Timeout in milliseconds
      },
      behaviour = {
        auto_suggestions = false, -- Experimental stage
        auto_set_highlight_group = true,
        auto_set_keymaps = true,
        auto_apply_diff_after_generation = false,
        support_paste_from_clipboard = false,
        minimize_diff = true, -- Whether to remove unchanged lines when applying a code block
      },

      mappings = {
        --- @class AvanteConflictMappings
        diff = {
          ours = "co",
          theirs = "ct",
          all_theirs = "ca",
          both = "cb",
          cursor = "cc",
          next = "]x",
          prev = "[x",
        },
        -- suggestion = {
        --   accept = "<M-l>",
        --   next = "<M-]>",
        --   prev = "<M-[>",
        --   dismiss = "<C-]>",
        -- },
        jump = {
          next = "]]",
          prev = "[[",
        },
        submit = {
          normal = "<CR>",
          insert = "<D-C-s>",
        },
        sidebar = {
          apply_all = "A",
          apply_cursor = "a",
          switch_windows = "<Tab>",
          reverse_switch_windows = "<S-Tab>",
        },
      },
      hints = { enabled = true },
      windows = {
        ---@type "right" | "left" | "top" | "bottom"
        position = "left", -- the position of the sidebar
        wrap = true, -- similar to vim.o.wrap
        width = 30, -- default % based on available width
        sidebar_header = {
          enabled = true, -- true, false to enable/disable the header
          align = "center", -- left, center, right for title
          rounded = true,
        },
        input = {
          prefix = "> ",
          height = 8, -- Height of the input window in vertical layout
        },
        edit = {
          border = "rounded",
          start_insert = true, -- Start insert mode when opening the edit window
        },
        ask = {
          floating = false, -- Open the 'AvanteAsk' prompt in a floating window
          start_insert = true, -- Start insert mode when opening the ask window
          border = "rounded",
          ---@type "ours" | "theirs"
          focus_on_apply = "ours", -- which diff to focus after applying
        },
      },
      highlights = {
        ---@type AvanteConflictHighlights
        diff = {
          current = "DiffText",
          incoming = "DiffAdd",
        },
      },
      --- @class AvanteConflictUserConfig
      diff = {
        autojump = true,
        ---@type string | fun(): any
        list_opener = "copen",
        --- Override the 'timeoutlen' setting while hovering over a diff (see :help timeoutlen).
        --- Helps to avoid entering operator-pending mode with diff mappings starting with `c`.
        --- Disable by setting to -1.
        override_timeoutlen = 500,
      },
      suggestion = {
        debounce = 600,
        throttle = 600,
      },
    },
    -- if you want to build from source then do `make BUILD_FROM_SOURCE=true`
    build = "make",
    -- build = "powershell -ExecutionPolicy Bypass -File Build.ps1 -BuildFromSource false" -- for windows
    dependencies = {
      "stevearc/dressing.nvim",
      "nvim-lua/plenary.nvim",
      "MunifTanjim/nui.nvim",
      --- The below dependencies are optional,
      "echasnovski/mini.pick", -- for file_selector provider mini.pick
      "nvim-telescope/telescope.nvim", -- for file_selector provider telescope
      "hrsh7th/nvim-cmp", -- autocompletion for avante commands and mentions
      "ibhagwan/fzf-lua", -- for file_selector provider fzf
      "nvim-tree/nvim-web-devicons", -- or echasnovski/mini.icons
      "zbirenbaum/copilot.lua", -- for providers='copilot'
      {
        -- support for image pasting
        "HakonHarnes/img-clip.nvim",
        event = "VeryLazy",
        opts = {
          -- recommended settings
          default = {
            embed_image_as_base64 = false,
            prompt_for_file_name = false,
            drag_and_drop = {
              insert_mode = true,
            },
            -- required for Windows users
            use_absolute_path = false,
          },
        },
      },
      {
        -- Make sure to set this up properly if you have lazy=true
        "MeanderingProgrammer/render-markdown.nvim",
        opts = {
          file_types = { "markdown", "Avante" },
        },
        ft = { "markdown", "Avante" },
      },
    },
  },

Environment

Neovim version:

nvim --version
NVIM v0.10.3
Build type: Release
LuaJIT 2.1.1734355927
Run "nvim -V1 -v" for more info

Distribution

NVChad

Platform

MacOS Sequoia 15.3 (24D60) on an M1 Macbook Pro

Repro