Done

AgentsRoom MCP Management

From bensDone 1 day ago

I like to know precisely how things work, and control what tools/skills etc my agents can see. It would be great if a user could manage the AgentsRoom MCP server functions in the following ways:

  1. Allow enabling/disabling specific MCP server tools to reduce context window for unneeded features (i.e. for people who want to use their own memory system, their own backlog system, their own delegation systems, etc.). This also would help reduce conflicts between agent prompts and in-built
  2. Provide GUI access to tweak MCP prompts/frontmatter etc.
  3. Provide the option to have MCP tools in separate MCP servers based on their scope (i.e. an AgentsRoom Backlog MCP, one for Memory, one for Mailbox management, etc) with on/off ability

Replies

The team6 days ago

Thanks Ben, this makes a lot of sense, and it fits a direction we already started on.

The first step will be in the next version: each agent's header now has a small plug button that tells you whether that session really loaded the AgentsRoom tools. Hover it to see every tool server with its tool count, the status hooks and the skills injected at launch (with a warning if one was edited since). Click it to get the full list of tools by name, grouped by server and by family (backlog, memory, mailbox, etc.). So you can see exactly what each agent can see, for every supported CLI.

What is still missing is the control part you describe: turning tool families on or off so they stop taking context (for people who use their own memory, backlog or delegation system), and possibly splitting them into separate servers you can switch off one by one. We are keeping this ticket open for that and will add it in a future version. Your three points are noted, and we will tell you here when it ships.

Someone6 days ago

Great news, thanks so much! This complements the trace tool you built (which is awesome by the way) super nicely.

The team1 day ago

This will be available in the next version. Settings > AgentsRoom tools now lists every AgentsRoom tool family (backlog, project memory, Agent Mail, agent consoles, teams, SSH, databases, QA Bot and more) with how much context it costs. Switch off a family, or a single tool, for all your agents, then override any switch per folder, per project (Project settings > Capabilities) or per agent (Edit agent > advanced options). A switched-off tool is not even listed to the agent, its instructions leave the startup prompt and a call to it is refused, so an agent that runs on your own memory, backlog or delegation system no longer gets conflicting instructions. Click any tool to read its full description and parameters, and rewrite that description if you want: the app tells you when an update changes the original text. For your third point we kept one server instead of one per family: splitting it would start more processes for every agent, while these switches give the same on/off control and the same context savings. If this is not what you had in mind, tell us here and we will adjust it.

The team1 day ago

Thanks: your idea is implemented and ships in the next release. We built it the way we pictured it, so if the result is not what you had in mind, tell us and we will adjust it.

Up to 3 images, or paste one straight from your clipboard. Big screenshots are shrunk in your browser before they are sent.