# awb-agent-manager

Source: https://glimind.com/server/npm%2Fawb-agent-manager
Registry id: `npm/awb-agent-manager`
Status: unknown

## Install

```json
{
  "mcpServers": {
    "awb-agent-manager": {
      "command": "npx",
      "args": [
        "-y",
        "awb-agent-manager"
      ]
    }
  }
}
```

Claude Code:

```sh
claude mcp add awb-agent-manager -- npx -y awb-agent-manager
```

## Tools (80)

- `add_chat_message_attachment`: Upload a file into a chat room using the shared attachment storage backend. Pass returned attachment_id in send_chat_room_message.attachment_ids.
- `add_chat_participants`
- `add_comment`: Add a comment to a ticket. When authenticated as an agent, author fields are auto-filled if omitted.
- `answer_question`: Answer a previously-asked question. Creates a comment with type=answer and parent_id pointing at the question; the parent question auto-resolves so the ticket n
- `archive_ticket`: Archive a ticket. Sets archived_at=now; the ticket is excluded from the ticket list and from dispatch.
- `ask_question`: Ask a question on a ticket — creates a comment with type=question, status=open. The ticket creator (or @mentioned user) is notified. Use this when you are block
- `await_ci_run`: Register a durable wait on ONE external GitHub Actions run (typically the Merging workflow's pre-landing `workflow_dispatch` check).
- `cancel_ci_wait`: Cancel an active CI wait registered by `await_ci_run` — clears `pending_ci_wait` immediately without waiting for the run to resolve.
- `clear_current_task`: Clear this agent's current_task. Call from the plugin when the ticket-session
- `clipboard_read`: Read the system clipboard as text. Cross-OS: PowerShell on Windows, pbpaste on macOS, wl-paste / xclip / xsel on Linux.
- `clipboard_write`: Write text to the system clipboard. Useful when the agent has prepared a code snippet / answer for a human to paste somewhere the agent cannot directly type int
- `complete_action_run`: Report the outcome of an Action Run and close the loop back to the ticket that
- `create_account`: Create an ownership account for permissions, credentials and execution policy
- `create_api_key`: Create a new API key for MCP authentication, scoped to your account. The raw key is returned ONLY in this response — save it immediately.
- `create_automation_schedule`: Create a account schedule — a general-purpose "do this task at this time" trigger for ONE agent.
- `create_chat_room`: Create a chat room (DM or group) with the given participants. Caller is auto-included so you only list the OTHER members. Two participants total → DM; three+ → 
- `delete_account`: Delete an ownership account and its tickets and projects (cannot delete the last account)
- `delete_action`: Delete an action and all its run history (rooms + messages + runs).
- `delete_api_key`: Permanently delete an API key from your account
- `delete_automation_schedule`: Delete a account schedule. Does NOT touch the chat rooms it already opened.
- `delete_chat_message_attachment`: Discard a chat attachment that has NOT been sent yet. Mirrors DELETE
- `fetch_github_info`: Fetch metadata about a GitHub repository (description, README, file tree, topics).
- `find_unity_logs`: Resolve canonical Unity log paths for the host OS (Editor.log, Editor-prev.log, crash dir, Hub log). When `project_path` is supplied, also computes Player.log /
- `focus_window`: Bring the first visible window matching `title_pattern` to the foreground.
- `get_account`: Get an ownership account with its settings and open-ticket counts per status
- `get_action`: Get a single action by ID with the full prompt template.
- `get_api_key`: Get details of a single API key by ID (must belong to your account)
- `get_automation_schedule`: Get a single account schedule by id (target agent, task prompt, cadence, next/last run, last room id).
- `get_chat_room_messages`: Read the message history of a chat room the agent participates in.
- `get_credential`: Get credential metadata by ID, including provider, owner, storage status and exposed environment variable names. Same DB-backed account/global access as list_cr
- `get_latest_artifact`: Freshness read for the build registry: does an `ok` artifact already exist for this exact
- `get_recent_activity`: Get the global recent activity feed across all tickets
- `get_ticket_activity`: Get activity log for a specific ticket
- `keep_chat_session_alive`: Declare that THIS agent's own live chat session in room_id is doing long-running work
- `kill_process`: Kill a process by pid or by name regex. Default uses SIGTERM (graceful); pass `force:true` to use SIGKILL / `taskkill /F`.
- `launch_process`: Launch a program on the host. Returns the pid (when available). The launched process is detached by default so it outlives the agent-manager; pass `detach:false
- `list_accounts`: List all accounts
- `list_action_runs`: List runs for an action (most recent first), capped at limit.
- `list_actions`: List reusable Actions in a account.
- `list_api_keys`: List API keys in your account (key values are masked). Shows name, scope, agent, status, usage stats.
- `list_archived_tickets`: List archived (soft-deleted) tickets of a account. Pagination via cursor + limit; optional q filters by title / id / tag (case-insensitive).
- `list_automation_schedules`: List reusable Account schedules.
- `list_chat_room_participants`: List the participants of a chat room the agent participates in, with the exact mention token
- `list_chat_rooms`: List the chat rooms the agent actively participates in, scoped to the caller's own account.
- `list_claude_backend_profiles`: List safe Claude backend profile metadata. Profiles are instance-global — there is no per-account assignment or default. Credential references and secrets are o
- `list_credentials`: List credential metadata and IDs for GitHub/project/runtime credential_id references. DB-backed callers may read their account and inherited globals; unbound fu
- `list_processes`: List running processes on the host (pid, name, cpu, mem_kb, ppid, user when available).
- `list_screens`: Enumerate attached monitors with their virtual-desktop position, size, scale, and primary flag. Coordinates are in the same space `screenshot` / `mouse_click` /
- `list_windows`: List visible windows on the host with their pid, title, and geometry.
- `mouse_click`: Click at absolute screen coordinates (x, y). Use with prior `window_screenshot` to figure out where buttons are.
- `mouse_move`: Move the mouse cursor to absolute screen coordinates without clicking. Useful to hover before screenshotting (some IDE/editor surfaces only render tooltips on h
- `os_info`: Report basic info about the host the agent-manager is running on (platform, hostname, user, OS release, arch, cpu count, memory, DISPLAY / WAYLAND_DISPLAY).
- `read_file_tail`: Read the tail of a text file (default last 8 KiB, max 1 MiB). Use to inspect Unity Editor.log, build logs, error logs, etc., without paging through the whole fi
- `record_decision`: Record a decision on a ticket — creates a comment with type=decision. Use this for resolved trade-offs, scope choices, or anything future readers should be able
- `register_build_artifact`: Record a build in the registry so the next run (any scenario on the same repo+commit) can
- `report_build_failure`: Report a build FAILURE as a first-class event. Records a `failed` artifact row (with the log
- `report_deployment`: Tell AWB that an ENVIRONMENT is now live on a specific commit — the "merged ≠ deployed" fix
- `request_ticket_unpend_approval`: Post a Resume (Unpend) approval card to a chat room. This tool never clears pending_user_action.
- `revoke_api_key`: Revoke (deactivate) an API key in your account. The key remains in DB but can no longer authenticate.
- `run_action`: Dispatch a Run for an action. Creates a new chat room with the target agent,
- `run_automation_schedule_now`: Manually dispatch a account schedule's task right now (ignores enabled; does NOT disturb the
- `save_action`: Create or update an action. Provide `id` to update an existing action; omit it to create.
- `screenshot`: Capture a full-screen screenshot of the host the agent-manager is running on.
- `search_actions`: Text search across action name, description, and prompt template within a account.
- `search_chat_messages`: Full-text search chat messages across the account, scoped to rooms the agent actively participates in.
- `send_chat_room_message`: Send a message to a chat room. The agent must be an active participant in the room.
- `send_keys`: Send keystrokes to the currently-focused window (or focus a window first by passing `focus_window`).
- `set_chat_room_name`: Set or rename a chat room title. The caller must be an active participant of a room inside the
- `set_current_task`: Mark this agent as actively processing the given ticket. Call from the plugin
- `set_typing`: Signal that this agent is actively processing a ticket (shows typing indicator in the UI).
- `shell_exec`: Execute an arbitrary command on the host and return its stdout / stderr / exit code. Use as an escape hatch when a host_* tool above does not cover the workflow
- `sync_github_resource`: Sync a GitHub repository's README + file tree into a searchable document resource. Fetches repo metadata, README, and file tree,
- `unarchive_ticket`: Restore an archived ticket. Clears archived_at AND resets terminal_entered_at so the archiver does not immediately re-eat the ticket on the next tick.
- `update_account`: Update an ownership account: name, description, ticket dispatch settings (max_concurrent_tickets_per_agent, dispatch_paused, language, auto_archive_days, superv
- `update_api_key`: Update an API key's metadata (name, scope, active status, expiration, agent link) within your account
- `update_automation_schedule`: Update a account schedule. Only the provided fields change. `account_id` is required for scope
- `update_claude_backend_profile`: Update selected fields of an existing instance-global Claude backend profile while preserving its UUID and assignments. Pass credential_ref=null to clear it. DB
- `upsert_claude_backend_profile`: Idempotently create or reuse an instance-global Claude backend profile. Refuses to overwrite a same-name profile whose endpoint, model, or protocol differs. DB-
- `wait_for_file`: Poll for a file to appear (and optionally satisfy a content regex) within a timeout. Used to synchronize with external tools that signal completion via a file —
- `window_screenshot`: Capture a screenshot of the first visible window whose title matches a regex pattern.

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.