The by-URL tools are the practical difference: paste the link a colleague dropped in chat and the server resolves it, rather than making you dig the page ID or issue key out of it first. Nothing here creates or edits, so the token can be scoped read-only and the worst a mistaken call does is fetch something.
A read-only server across two Atlassian products: Jira issues and projects, Confluence pages and spaces. Every tool fetches; none of them write.
- A Jira issue by key, such as PROJ-123
- A Jira issue from a pasted URL, including board URLs that carry the issue in a query parameter
- Jira issues matching a JQL query
- The Jira projects your token can see
- A Confluence page by ID
- A Confluence page from a pasted URL, in the spaces, display and viewpage.action forms
- Confluence pages matching a search
- The Confluence spaces your token can see
ATLASSIAN_DOMAIN and ATLASSIAN_EMAIL, plus API tokens — ATLASSIAN_CONFLUENCE_TOKEN and ATLASSIAN_JIRA_TOKEN separately, or a single combined token with access to both. Read-only scopes are enough: read:content:confluence and read:content-details:confluence on the Confluence side.
Build from source — clone the repository and build it, then point your client at the binary
