api-reference

name	api-reference
description	This skill should be used when the user asks for "Jira API endpoints", "Confluence API endpoints", "Jira/Confluence REST API", "Atlassian curl examples", "the exact method/parameters" for any Jira or Confluence resource, or needs HTTP details for issues, JQL search, comments, worklogs, attachments, links, projects, versions, components, fields, screens, workflows, statuses, users, groups, permissions, schemes, dashboards, filters, plans; or Confluence pages, blog posts, spaces, comments, attachments, labels, content properties. Indexes the full per-domain endpoint catalog plus the bundled OpenAPI specs.
disable-model-invocation	true

Atlassian REST API Reference (Jira v3 + Confluence v2)

Complete catalog for the Jira Cloud platform REST API v3 and the Confluence Cloud REST API v2, split by domain. Official docs: https://developer.atlassian.com/cloud/jira/platform/rest/v3/ and https://developer.atlassian.com/cloud/confluence/rest/v2/

The exhaustive source of truth is the two bundled OpenAPI specs:

• references/jira-openapi-v3.json — 420 paths, 99 tags (OpenAPI 3.0.1).
• references/confluence-openapi-v2.json — 147 paths, 29 tags (OpenAPI 3.0.3).

The curated references/jira/*.md and references/confluence/*.md files below cover the common operations in plain, copy-pasteable form.

Load the setup skill first for authentication and the global conventions (Basic auth, headers, ADF/storage bodies, Jira startAt/maxResults vs Confluence cursor pagination, accountId) — those rules apply to every method here and are not repeated in each file.

Global recap

• Jira base: ${ATLASSIAN_SITE_URL%/}/rest/api/3. Confluence base: ${ATLASSIAN_SITE_URL%/}/wiki/api/v2.
• Auth: -u "${ATLASSIAN_EMAIL}:${ATLASSIAN_API_TOKEN}", -H "Accept: application/json" (+ -H "Content-Type: application/json" on writes).
• Jira rich text = ADF JSON (description, comment body). Confluence body = {representation, value}; update needs the next version.number.
• Jira users = accountId (resolve via GET /user/search). Paths in the Jira files below are written relative to the /rest/api/3 base; Confluence paths are relative to the /wiki/api/v2 base.

Jira domain index

Domain	File	Covers
Issues	references/jira/issues.md	createIssue, getIssue, editIssue, deleteIssue, assignIssue, getTransitions/doTransition, createmeta/editmeta, notify, changelog, archive/unarchive, bulk create/edit/move/delete/transition/watch
Search & JQL	references/jira/search-jql.md	search/jql (GET+POST), search/approximate-count, jql/parse, jql/autocompletedata, jql/match, issue/picker, JQL syntax primer
Comments, worklogs & properties	references/jira/comments-worklogs.md	issue comments CRUD, worklogs CRUD + move + deleted/updated, issue properties
Attachments & links	references/jira/attachments-links.md	add/get/delete attachments (multipart), issue links + link types, remote links
Projects, versions & components	references/jira/projects-versions-components.md	projects CRUD + search + archive/restore, versions, components, roles & actors, categories, features, templates
Fields & screens	references/jira/fields-screens.md	issue fields, custom field contexts & options, field configurations (+schemes), screens, screen tabs, screen schemes, issue type screen schemes
Workflows, types & statuses	references/jira/workflows-types-statuses.md	workflows (read/create/update/search), workflow schemes (+drafts), statuses, issue types, issue type schemes, transition rules
Users & groups	references/jira/users-groups.md	users CRUD, user search (accountId), groups + members, myself + preferences, avatars
Permissions & schemes	references/jira/permissions-schemes.md	permission schemes + grants, my/all permissions, issue security schemes & levels, notification schemes, priorities + priority schemes, resolutions
Dashboards & filters	references/jira/dashboards-filters.md	dashboards + gadgets + item properties, filters CRUD + columns + favourite + owner, filter sharing
Plans & teams	references/jira/plans-teams.md	Advanced Roadmaps plans CRUD + archive/duplicate/trash, teams in plan (atlassian + plan-only)

Confluence domain index

Domain	File	Covers
Pages & blog posts	references/confluence/pages-blogposts.md	pages CRUD (versioned), blog posts, body-format, ancestors/children/descendants, custom content, whiteboards, databases, folders, smart links
Spaces	references/confluence/spaces.md	spaces list/get/create, space properties, space permissions, space roles
Comments & attachments	references/confluence/comments-attachments.md	footer + inline comments CRUD, attachments, versions, likes, tasks, operations
Labels & content properties	references/confluence/labels-content-properties.md	labels, content properties, classification levels, data policies, redactions, admin key, app properties

How to use a reference file

Each file lists endpoints as METHOD /path (relative to the product base) with purpose and key fields. To run one, wrap it with auth and jq:

# Jira example
curl -s -u "${ATLASSIAN_EMAIL}:${ATLASSIAN_API_TOKEN}" -H "Accept: application/json" \
  "${ATLASSIAN_SITE_URL%/}/rest/api/3/<path>" | jq '.'
# Confluence example
curl -s -u "${ATLASSIAN_EMAIL}:${ATLASSIAN_API_TOKEN}" -H "Accept: application/json" \
  "${ATLASSIAN_SITE_URL%/}/wiki/api/v2/<path>" | jq '.'

For the full request/response schema of any operation, grep the bundled spec by operationId:

grep -n '"operationId": "createIssue"' references/jira-openapi-v3.json
grep -n '"operationId": "createPage"'  references/confluence-openapi-v2.json

Examples

User: "What's the endpoint to create a Jira issue?" → Open `references/jira/issues.md`. `POST /issue` with `{"fields":{"project":{"key":"PROJ"},"issuetype":{"name":"Task"},"summary":"…","description":}}`. `description` must be ADF JSON, not a string. User: "How do I search issues with JQL?" → Open `references/jira/search-jql.md`. `POST /search/jql` with `{"jql":"project = PROJ ORDER BY created DESC","fields":["summary","status"],"maxResults":50}`. Page with the returned `nextPageToken`. User: "Give me the curl to update a Confluence page." → Open `references/confluence/pages-blogposts.md`. `GET /pages/{id}?body-format=storage` to read the current `version.number`, then `PUT /pages/{id}` with `{"id","status":"current","title","version":{"number":},"body":{"representation":"storage","value":"

…

"}}`.