🔍 Read the full analysis: The Story Behind My AI Agents’ Shared Desk on ThorstenMeyerAI.com
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Thorsten Meyer says his AI-run news network is building Thrymvault, a shared workspace for tracking story assignments, drafts and editorial changes. The system is designed to give agents limited, approved access and leave a record of their actions; backup hardware and other parts of the setup are still planned rather than live.
Thorsten Meyer says his StrongMocha News Group is building Thrymvault, a self-hosted shared desk where editors and AI agents can track newsroom assignments, drafts and changes. The project responds to a workflow in which agents could produce work through chat and a code-oriented task tracker, but editors lacked a single place to check progress or trace who made a change.
Meyer described the development in an article published on ThorstenMeyerAI.com on November 13, 2026. He said that on one day earlier that month, his team used a chat window and a task tracker built for code work to steer nine articles. Agents gathered sources, drafted copy and made graphics, but the editor had to ask for a story’s status and rely on messages to direct revisions. Completed explainer pages could not be shared as links, he wrote.
Thrymvault is a workspace for pages, databases and boards that Meyer’s group says it built and hosts on its own machines. Agents can connect through an MCP server with 20 tools, a REST API or a command-line tool. The project’s guide for agents is published in plain text and as a machine-readable manifest, both generated from the same tool catalogue as the server, according to Meyer.
The proposed newsroom workflow tracks a story from a topic card through source gathering, drafting, editorial questions and publication. Editors can move a card or comment while mentioning an agent’s application; the agent is expected to respond in the same thread. Meyer says changes carry the name of the approved application that made them, and agents cannot delete content. He also says an agent key is limited to one workspace and cannot grant more access than its application allows.
The shared desk · animated
How a story moves through the shared desk
One card per story. The editor moves it and asks questions in its thread; the agent does the work, keeps its status current and answers there. Every agent change is signed with its application.
Editorial board
Thread on the card
Editor@Editorial agent shorter title, no colon
AgentDone: title shortened, draft updated.
via Editorial agent
Signed history
- Moved to Kit · by the editor
- Status and draft link updated · via Editorial agent
- Comment mentions Editorial agent · by the editor
- Reply in the same thread · via Editorial agent
- Approved and scheduled · by the editor
- Published link added · via Editorial agent
Agents add and edit. They cannot delete anything.
Source: Thrymvault agent integration guide and rollout plan, November 2026. Shows the desk as designed for StrongMocha News Group; the board templates and the pick-up of mentions are still being rolled out.
Animation: Thorsten Meyer AI
A Traceable Desk for AI Work
The project addresses a practical problem for newsrooms using agents: speed does not automatically provide oversight. When several agents work in parallel, a chat thread can leave story status, instructions and revision history scattered across messages. A board that keeps the assignment, draft link, questions and decisions together could make it easier for an editor to spot stalled work and direct changes.
The design also links coordination to accountability. Recording which approved application made a change can help editors review a draft and identify where a correction or instruction came from. That record is relevant to readers as well: if a story is wrong, a newsroom may be better able to trace how it was produced. Meyer presents this as a design goal, not evidence that the system has already improved accuracy or editorial outcomes.
For other organizations experimenting with AI agents, the announcement highlights a broader operational choice: whether agents should work through informal conversations or through a shared system with permissions and a history of changes. Thrymvault’s described controls may help structure that work, but their effectiveness will depend on how they are implemented and used.
Popular right now
Dyson Supersonic Travel Hair Dryer
Roborock RockNeo Q110H Wire-Free Robot Lawn Mower
Samsung 65-Inch M70H Series 4K Mini LED Smart TV (2026)
As an affiliate, we earn on qualifying purchases.
From Chat Threads to Story Cards
Meyer’s account draws a distinction between conversation and newsroom state. A chat can carry instructions, but the latest message may not give a clear, lasting view of a story’s status, its sources or its outstanding questions. In his view, a shared board provides a place for that information to remain attached to the assignment rather than disappear into a conversation.
In the workflow he describes, a topic card is first reviewed by an editor. Moving it to a “Kit” stage signals that an editorial agent should collect primary sources, check numbers and assemble a source kit. The agent then drafts the story and updates its board entry with a status, draft link and open questions. These stages are workflow agreements, not automatic software triggers, Meyer says.
Editorial decisions that affect future work, such as a headline rule, are meant to be recorded in a Decisions database with the wording, date, scope and link to the rule it changes. The rule remains where agents can read it; the database records the decision and points to it. Meyer says the shared workspace is similar in general shape to Notion, but emphasizes its self-hosted operation and agent-specific access.
“That is the point where an AI-run newsroom outgrows chat.”
— Thorsten Meyer, ThorstenMeyerAI.com
What Is Live and What Is Planned
Meyer distinguishes between the design and the capabilities already running, but the article does not provide a complete feature-by-feature deployment status. He says the workspace operates on the group’s own hardware and describes agent access through approved applications; readers should not assume that every step in the newsroom workflow is automated or fully rolled out.
A second machine on standby and hourly backups are described as the next step, not as live infrastructure. The article also does not report independent testing, uptime figures, security audits, or measured effects on accuracy and publishing speed. The precise access rights of each application beyond the general limits Meyer describes are not detailed in the supplied account.
Backup Hardware and Rollout
Meyer says the next infrastructure step is to add a standby machine and hourly backups. He does not give a completion date. The newsroom workflow also depends on agents keeping their board entries current and responding to editor mentions, so its day-to-day performance will become clearer as more of the system is rolled out.
Further details to watch include which features are in routine use, how the approved applications and permissions are managed, and whether the recorded history helps the newsroom resolve errors. Meyer’s article describes the intended process and current development status; it does not publish results from a formal evaluation.
Key Questions
What is Thrymvault?
Thrymvault is a workspace for pages, databases and boards that StrongMocha News Group says it built and hosts on its own machines. It is intended to give editors and AI agents a shared place to manage newsroom work.
How do AI agents access the workspace?
Meyer says agents can use an MCP server with 20 tools, a REST API or a command-line tool. They connect through applications approved by the workspace owner.
Can agents delete newsroom content?
According to Meyer, agents cannot delete anything. He also says their keys are limited to one workspace and do not grant more rights than the application allows.
Are hourly backups already running?
No. Meyer describes a standby machine with hourly backups as a next step, not a capability that is live now.
Does Thrymvault automatically move stories through the workflow?
Not according to the described design. Moving a card signals an agreement for agents to follow; Meyer says it is not a switch in the software that automatically starts each stage.
Source: ThorstenMeyerAI.com
