About TemplatePoint
Reusable text without giving up control
TemplatePoint is a shared template platform for teams that repeat important wording. Maintain the parts that should stay consistent, turn the parts that vary into structured fields, and let people fill in only what changes.
The product
The wording stays curated. The details stay flexible.
Copying last week's reply, agreement, or command feels quick, but it quietly creates drift: old details survive, approved wording changes, and every person ends up maintaining a different version.
TemplatePoint gives a team one maintained source for the text that matters. Editors control the structure and wording; everyone else can fill, reuse, and export it without rewriting the parts that should remain stable.
One maintained source
Update the template once so the team stops collecting slightly different personal copies.
Fields, not find-and-replace
Names, dates, numbers, choices, and images become clear inputs inside the text.
Roles and history
Separate editing from everyday use and keep a record of how important content changes.
How it is built
AI-assisted, human-accountable
TemplatePoint is designed, built, and operated by Marius Solheim as a one-person, bootstrapped product.
I use AI extensively for research, implementation, testing, and review. It helps one developer move faster, but it does not own the product direction, validate its own findings, or decide what reaches production. Those decisions and that responsibility remain mine.
Security
Security is part of the release process
TemplatePoint stores team content across organizations and workspaces. Access boundaries and release discipline therefore have to be part of how the product is built, not an afterthought added at the end.
Access enforced in the database
Tenant access is enforced with Postgres Row Level Security as well as application checks, so hiding a control in the interface is never treated as the security boundary.
Boundaries tested as real users
Authorization tests exercise separate organizations, lower-privilege roles, anonymous requests, mismatched resource identifiers, and attempts to cross ownership boundaries.
Review required before release
Security-sensitive changes must use a reviewable pull request, receive an independent security and correctness review, resolve confirmed material findings, and record the relevant verification before merging to the production branch.
Fixed weaknesses become regression tests
When a weakness is remediated, its reproducer is retained and turned into a permanent test that requires the unsafe action to be denied and verifies that data remains unchanged.
Specific controls, not a certification claim
TemplatePoint does not currently claim a formal security certification or pretend that every safeguard is fully automated. The promise is narrower: describe the controls accurately, document gaps, and strengthen enforcement instead of hiding behind broad claims.
Where it started
A customer-support problem, not a software trend
The idea began while I worked in IKEA customer support. Routine cases could be automated, but the moment a delayed order, claim, or deviation needed a human response, every case was different.
The practical tool was a collection of sticky notes with reusable text and XXX wherever a name, order number, or case number belonged. It worked until replacing one XXX meant carefully editing inside a sentence without damaging anything around it.
That frustration became the first prototype: stable text with real inline fields. The product has grown into shared workspaces, rich documents, AI-assisted drafting, roles, and version history, but the central idea is still the same.
Dear Customer name,
We have tried to contact you regarding case Case number.
Best regards, Customer support
Direct accountability
One product, one accountable owner
There is no distance between product decisions, security responsibility, and the person you can contact. If you are considering TemplatePoint for your team, you will speak directly with the person building it.