🔍 Read the full analysis: I Put My AI Agents At One Desk—Here’s Where Thrymvault Fits on ThorstenMeyerAI.com
Get the little things that make your day delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Thorsten Meyer says his AI-run newsroom is adopting Thrymvault, a self-hosted workspace where editors and agents can track stories, drafts and decisions. Its documented controls include approved applications, changes attributed to the application that made them, and no agent deletion rights. A standby machine and hourly backups are planned, not yet live.
Thorsten Meyer says his AI-run publishing operation is adopting Thrymvault, a self-hosted workspace intended to give editors and AI agents one place to track stories, drafts and decisions. In an account published November 13, he describes controls including approved agent applications, changes attributed to the application that made them, and a rule that agents cannot delete content; he says a standby machine and hourly backups are still planned rather than live.
Meyer says that on one day earlier in November, his team moved nine articles through a chat window and a task tracker designed for code work. Agents gathered sources, drafted stories and prepared graphics, but the editor had to ask for progress, relay headline changes through chat and find another way to share a completed explainer. He presents those problems as the reason the newsroom needs a shared desk rather than relying on conversations alone.
Thrymvault is described as a workspace for pages, databases and boards, built and hosted on the publisher’s own machines. According to Meyer’s account and the cited agent guide, agents can work through an MCP server with 20 tools, a REST API or a command-line tool. The agent guide is published as plain text and as a machine-readable manifest, both generated from the same tool catalogue as the server.
In the proposed editorial workflow, a story begins as a card on a topic board. An editor decides whether to move it forward; the editorial agent then prepares a source kit, drafts the story and updates its status and report on the board. The editor can request changes in a card’s comment thread by mentioning the relevant agent application. The intended final record includes the published link, while editorial decisions, such as a new headline rule, are logged in a separate database.
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 Agent Work
The practical issue is editorial oversight. When several agents handle sourcing, drafting and graphics in parallel, a chat transcript can make it difficult to see which version is current, who requested a change or whether an instruction was followed. A board that keeps status, comments and reports alongside a story could make those handoffs easier to inspect.
Meyer also links the setup to accountability for published work: a record of who changed what and which application made the change could help an editor investigate and correct errors. That is a description of the system’s intended value, not evidence in the source that it has already reduced mistakes or improved publication quality. For other organizations using agents, the account offers a model centered on explicit approvals and shared records rather than informal chat alone.
Products Worth Considering
Hyperice Hypervolt Go 3 Percussion Massage Gun
Oral-B iO5 Rechargeable Electric Toothbrush with 5 Cleaning Modes and Travel Case, White
Garmin Venu 3S GPS Fitness Smartwatch, 41 mm, French Gray
As an affiliate, we earn on qualifying purchases.
From Chat Threads to Story Cards
Meyer writes that StrongMocha News Group, his network of niche and news sites, operates with AI agents. He says chat worked for issuing instructions but provided a weak shared view of a newsroom’s ongoing work: story state depended on recent messages, and important directions could scroll out of view. His comparison is with a newsroom desk, where each story has a status, attached editorial notes and rules about who can change it.
Thrymvault is presented as a product with a familiar pages-and-databases format, but the article focuses on its agent access model. Access is granted through applications approved by a workspace owner. Meyer says each machine has a key, and changes are marked as made “via” the relevant application. The account says an agent key carries the rights of the person who created it, within one workspace, and cannot exceed the application’s permissions. These are descriptions from the product’s guide and development material as cited in the article, not an independent security audit.
“That is the point where an AI-run newsroom outgrows chat.”
— Thorsten Meyer, describing the newsroom’s workflow
What Is Live and What Is Planned
The article distinguishes documented agent controls from infrastructure that has not yet been deployed. Meyer says the workspace runs on the publisher’s own hardware, while a second machine on standby and hourly backups are next steps, not live features. It does not provide a deployment date for those additions.
The account describes how the newsroom workflow is designed, but it does not specify how many of the nine articles were handled through Thrymvault itself, whether every proposed board and database is already in routine use, or how well the process performs over time. It also offers no independent review of the access controls, security, reliability or effect on error rates. The source says agents cannot delete content, but does not detail how the system handles mistaken edits, outages or recovery beyond the planned backups.
Standby Hardware and Workflow Rollout
Meyer identifies the standby machine and hourly backups as the next infrastructure work. He does not give dates or a rollout schedule. The described newsroom process also calls for agents to maintain story status, attach draft reports to cards and respond to editorial requests in threads; the article does not set out a timetable for making each step standard practice.
The next useful test will be whether the workspace produces a reliable record across real assignments, from initial sourcing through publication, while keeping access limited to approved applications. Readers should distinguish that future evaluation from the claims currently supported by the account: the system’s described design and controls, and the publisher’s report of a nine-article day coordinated through chat and a code-oriented tracker.
Key Questions
What is Thrymvault?
It is a self-hosted workspace for pages, databases and boards. Meyer describes it as a shared desk where editors and AI agents can track story work and decisions.
How do AI agents access the workspace?
The article says agents can connect through an MCP server with 20 tools, a REST API or a command-line tool. Access is through applications approved by the workspace owner.
Can agents delete stories or other content?
According to the agent guide as described by Meyer, agents cannot delete anything. The article does not provide an independent security assessment of that control.
Are hourly backups already running?
No. Meyer says hourly backups and a second standby machine are planned next steps and are not live yet.
Does moving a story card automatically start an agent?
Not according to the article. Meyer says moving a card is an agreement agents are expected to follow, not a software switch that automatically triggers the work.
Source: ThorstenMeyerAI.com
