Skip to main content
Glama
Sign In
Keyhole – Local files for ChatGPT

Keyhole – Local files for ChatGPT

by L-Jovi
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Related Servers

Alternatives to Keyhole – Local files for ChatGPT

No user-submitted related servers found.

    Related Servers

    • WebCodex-Bare

      Tim-Devil
      A
      license
      Not graded
      quality
      C
      maintenance
      Enables ChatGPT to browse, read, search, create, and edit files, run commands and Python, and use Git within configured local workspaces, with permission controls, background job management, and tunnel-based access.
      MIT

    • ChatGPT MCP Bridge

      AresQQQQQ
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables ChatGPT to read and edit local project files, inspect Git changes, and run approved development scripts through a secure MCP tunnel, with optional Codex Desktop integration.
      31 npm
      8
      Apache 2.0
    • PC FileBridge

      PC FileBridgeofficial

      gexiro-global
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables ChatGPT, Codex, and compatible MCP clients to read bounded UTF-8 text and create files or directories in operator-selected folders, with strict protections against overwrites, deletes, and path escapes.
      Apache 2.0

    • Local Computer Bridge

      szkeli
      A
      license
      A
      quality
      C
      maintenance
      Enables ChatGPT to read and edit local text files, apply precise string replacements, and run Bash commands on a user's own macOS or Linux machine through a secure local tunnel without exposing public ports.
      4
      MIT

    • Local Supervisor Bridge

      VITASID57
      A
      license
      Not graded
      quality
      C
      maintenance
      Enables ChatGPT web to safely read and modify only explicitly allowed local project files through OpenAI Secure MCP Tunnel, including git operations, file edits, and running project scripts, while enforcing strict security boundaries.
      29
      MIT
    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A4.2/5.0

    Scored across 12 tools

    Disambiguation5/5

    Every tool has a unique, well-defined purpose with no overlap. For example, list_directory enumerates entries, search_files performs content searches, and read_file retrieves file contents with hashing options. Even similar operations like write_file (full replace) and apply_text_patch (partial edits) are clearly separated.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun pattern in snake_case: list_workspaces, search_files, read_file, apply_text_patch, create_directory, copy_file, etc. There are no deviations in style or grammar, making the API predictable and easy to navigate.

    Tool Count5/5

    12 tools is ideal for a file management server. The count is sufficient to cover the full lifecycle of file operations—listing, searching, reading, writing, patching, copying, moving, deleting, and undoing changes—without redundancy or bloat.

    Completeness5/5

    The tool surface covers the entire file management domain: discovery (list_workspaces, list_directory), search (search_files), read (read_file with multiple formats), write (write_file, apply_text_patch), directory operations (create_directory), file manipulation (copy_file, move_file, delete_file), and change management (list_changes, restore_change). No critical operations are missing for the stated purpose.

    Maintenance

    ActivityNo data
    ResponsivenessOutdated data

    Resources