REST API

Gisia exposes machine-readable skill files at well-known URLs. Instead of reading static API documentation, your AI bot fetches the skill file directly from your live instance — so the URLs, IDs, and scopes it receives are always accurate and up to date.

Why Fetch Instead of Document

Skill files are intentionally not reproduced here as static examples. There are two reasons:

• Instance-specific values — base URLs, project IDs, and token scopes differ between installations. A hardcoded example would require manual substitution and is error-prone.
• Always current — when endpoints or fields change, the skill file updates automatically. A bot that fetches the file directly always has the correct instructions without any doc maintenance.

The recommended pattern is to give your bot the skill URL and let it read the live file rather than pasting example curl commands from documentation.

Available Skills

• /-/users/skill.md — current authenticated user and links to all user-level skills
• /:namespace/:project/-/skill.md — project overview and links to all project-level skills

Both files link to their sub-skills, so your bot can navigate the full API from these two entry points.

To view these files in a browser, you need to be logged in.

How to Use

Give your bot one of the skill URLs above along with your personal access token, then ask it to download all the listed skills.

Read /-/users/skill.md using PRIVATE-TOKEN: YOUR_PERSONAL_ACCESS_TOKEN

Read the project skill and download all skills by running:
curl -s https://your-instance/:namespace/:project/-/skill.md -H "PRIVATE-TOKEN: YOUR_PERSONAL_ACCESS_TOKEN"