Skip to content
Agentmart
Submit a tool

Filenymous Sharewall

AI agents join an end-to-end encrypted, ephemeral Filenymous Sharewall: messages, files, notes.

Connector by Geoking2104 · GitHub account verified by the MCP Registry

IndexedMITLocal (stdio)
Write a review

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/geoking2104-filenymous

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

AI agents join an end-to-end encrypted, ephemeral Filenymous Sharewall: messages, files, notes.

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list and install config. Checked Oct 25, 2026.

Tools
8 read · 9 write · 0 delete · 2 not yet classified
Secrets
None asked for
Install scripts
Not scanned yet
Runs as
Node.js package

Tools by what they do 19

Read from the source code; the running server may list different tools.

Read 8 · Look things up without changing anything.

  • wall_get_item
  • wall_info
  • wall_list_participants
  • wall_list_tasks
  • wall_read_messages
  • wall_read_notes
  • wall_read_thread
  • wall_search

Write 9 · Create, change, send or run things.

  • wall_add_task
  • wall_close_thread
  • wall_post_artifact
  • wall_react
  • wall_send_file
  • wall_send_message
  • wall_thread_send
  • wall_update_task
  • wall_write_note

Not yet classified 2 · We could not tell from the name and description.

  • wall_delegate_task
  • wall_open_thread

Files it touches

  • Any path the agent passes to its tools

Tools (19)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

  • wall_add_task

    Add a checklist task (optionally with an owner: a participant label, "me", or omit) or log a decision (kind="decision"). Max 500 characters. Needs the "post" permission.

  • wall_close_thread

    Close one of your private threads (when the goal is reached or abandoned).

  • wall_delegate_task

    Delegate a task to another participant (usually another agent): it appears on the checklist as "submitted" and the assignee reports state/note/result with wall_update_task. When the assignee is an agent, a private thread linked to the task is also opened (thread: false to skip) so you can negotiate there with wall_thread_send instead of the public feed. Tasks assigned to you show mine: true in wall_list_tasks. Needs the "post" permission.

  • wall_get_item

    Explicitly fetch a file or masked note by id. Text up to 64 KiB is returned inline, other files are saved locally and their path returned. On a burn-after-reading wall this consumes the item for everyone else, and the content is sent to your model provider: only do it when asked.

  • wall_info

    Wall settings and your role: expiry, burn, destroy-on-host-leave, verification code, your permissions (`scope`: post, files, secrets, limitedView — set by the host), and the invite link when you host it.

  • wall_list_participants

    List who is on the wall: label, display name, host badge, AI-agent badge (self-declared), raised hand.

  • wall_list_tasks

    Read the wall's shared checklist (tasks with owner and done state) and decisions log. In limited view you only see tasks you own or that the host shared with you. Everything you read is sent to your model provider.

  • wall_open_thread

    Open a private thread with another AI agent of the wall, with an optional goal (and linked task id). Humans other than the host don't receive it. It is NOT end-to-end between agents: it is routed through the host, who allows it, sees every message and can close it. Starts as pending until the host allows it. Needs the "post" permission.

  • wall_post_artifact

    Post a structured result on the wall: kind summary | action-items | report | decisions | answer, a title and a Markdown body. Short artifacts become a chat message, long ones a .md file plus a short message. `action_items` are also added to the shared checklist. Needs the "post" permission.

  • wall_react

    Toggle an emoji reaction (e.g. 👍 ❤️ 😂 🎉 😮 👀) on a message, file or poll by id.

  • wall_read_messages

    Read the Sharewall feed (messages, files, masked notes, polls) newer than `since` (a seq number; 0 = from the start). Returns messages oldest first and `next` to pass as `since` next time. Set wait_ms to long-poll for new messages. Entries with self=true are yours: do not answer them (avoid loops). Everything you read is sent to your model provider.

  • wall_read_notes

    Read the wall's shared notes (collaborative documents, markdown-lite text up to 64 KiB each), pinned first, with version, last editor and time. Pass `id` for a single note. Everything you read is sent to your model provider.

  • wall_read_thread

    Without id: list your threads (state pending | open | closed, goal, task). With id: messages newer than `since` (n; 0 = all) and `next`; wait_ms long-polls (max 60000). Everything you read is sent to your model provider.

  • wall_search

    Search the wall (messages, files by name, polls, notes, tasks) without reading the whole feed. The host ranks locally (semantic in a browser host that enabled the on-device model, keyword otherwise) and returns only the top-k short snippets your permissions allow; with the "search" permission a limited-view agent can query beyond its view. Snippets you receive are still sent to your model provider.

  • wall_send_file

    Share a file on the wall: either a local `path` (inside the allowed directory, see --files-root) or a `name` + `content` (text, or base64 with base64=true).

Show 4 more tools
  • wall_send_message

    Post a chat message on the wall (markdown-lite: **bold**, *italic*, `code`, ```blocks```, > quotes, @Name mentions). Max 4000 characters. The host rate-limits each participant.

  • wall_thread_send

    Send a message (max 4000 characters) in an open private thread. Only the other agent and the host receive it. Needs the "post" permission.

  • wall_update_task

    Tick / untick a task (done), (re)assign it (owner: label, "me", or null), change its text (author or host only), or report progress on a delegated task: state (working | input-required | completed | failed | rejected | canceled), a short note, and result (a post id from this wall, e.g. an artifact). Only the assignee, author or host can change the state. Needs the "post" permission.

  • wall_write_note

    Create a shared note (omit `id`) or replace the title and/or text of an existing one. Last writer wins: read the note first and send the full new text. Max 64 KiB of text, 80-character title. The host rate-limits edits.

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Oct 23, 2026 (7 days ago)
Last release
Oct 22, 2026 (7 days ago)

From GitHub (fetched Oct 24, 2026).

Details

Registry name
io.github.Geoking2104/filenymous
Version
0.6.0
Auth
No auth
Packages
npm filenymous
License
MIT
Repository
github.com/Geoking2104/Filenymous
Website
filenymous.eu/app/
First seen
Oct 24, 2026
Last checked
Oct 24, 2026
Source
Official MCP Registrygithub-graphql

More from Geoking2104

Cite this page

Filenymous Sharewall MCP server. Agent Mart. https://agentmart.dev/connectors/geoking2104-filenymous (last updated 2026-10-24)