Run Tekio Delivery on your own machine

Two commands and a Jira API token. It runs entirely inside your network — nothing is sent anywhere. Budget about twenty minutes, most of it in the configuration wizard.

What you need

Docker with Compose — Docker Desktop on Windows or macOS, Docker Engine on Linux
2 GB of RAM and about 5 GB of disk for the cache
A Jira Cloud account that can read the projects you want to measure
No Jira admin rights — read access is enough

No Jira at all? You can feed it CSV files instead — see step 5.

Pull the image and start it

Create a folder, drop in the compose file, and start. The image is public — no login, no account.

# 1 — compose file
curl -O https://delivery.tekio.app/docker-compose.yml

# 2 — a secret, so your Jira token is encrypted at rest
echo "TEKIO_SECRET_KEY=$(openssl rand -hex 32)" > .env

# 3 — start
docker compose up -d

Open http://localhost:3000. First start pulls about 1.2 GB, so give it a minute.

Connect your Jira

In Jira, create an API token at id.atlassian.com → Security → API tokens. Then in Tekio Delivery, go to Settings → Connection and enter three things:

Field	Example
Jira URL	https://yourcompany.atlassian.net
Email	the account the token belongs to
API token	the token you just created

The token is encrypted with your own secret and stored in the container's volume. It never leaves your machine — without TEKIO_SECRET_KEY set, the app refuses to store it at all rather than writing it in the clear.

Onboard your first project

The wizard reads your Jira and proposes a mapping: issue types, the fields that hold team and story points, and which statuses count as in-progress. It ranks each candidate with evidence — how often the field is filled, what values it actually holds — so you're confirming, not guessing.

It won't let you finish until the configuration validates against your real issues. Pick Scrum or Kanban for a single team; that's the free tier, with no time limit.

SAFe mode — several teams on one programme, with the cross-team rollup — is what the licence unlocks. →

No Jira? Load CSV files instead

If your Jira is locked down, or you're on Azure DevOps or something else entirely, you can feed the product three CSV files instead of an API connection.

File	One row per	Columns
issues.csv	issue	key, type, status, points, team, epic, created, resolved
transitions.csv	status change	key, from, to, timestamp
sprints.csv	sprint membership	sprint, key, start, end, added_at, removed_at

Download the template pack — a filled example (Northwind Maritime) beside an empty file with the same columns. Open the filled one first; it's clearer than any schema documentation.

transitions.csv matters most. It's the status history, and it's where cycle time, flow efficiency, spillover and committed-versus-delivered come from. Without it you get counts and current state — a much poorer picture.

Updating

Atlassian changes the Jira API regularly, so an instance that stops getting updates will eventually break. Updating is two commands, and your configuration and cache are preserved in the volume.

docker compose pull
docker compose up -d

If something goes wrong

Symptom	Usually
Blank page on :3000	Still starting. Check docker compose logs -f.
“Cannot store token”	TEKIO_SECRET_KEY isn't set in .env.
401 from Jira	Email and token must belong to the same account.
Wizard finds no fields	The account can't see the project. Check permissions.
Charts are empty	First sync is still running — it reads changelogs, which takes a few minutes.

The free tier comes with no support — that's the trade for free. Bugs are welcome as GitHub issues, and if you'd rather not do any of this yourself, I'll install it for you.