Skip to main content
Glama
Sign In
deenesjakoruzh
@ama-work/mcp-atlassian

@ama-work/mcp-atlassian

by Ask-Me-Anything-dot-work
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Remote

ama-mcp-atlassian

Standalone TypeScript MCP server wrapping the Atlassian REST API (Jira + Confluence). Works standalone via env vars or registry-integrated via ama-mcp-registry.

Install

npx -y @ama-work/mcp-atlassian
# or
bunx @ama-work/mcp-atlassian

Related MCP server: mcp-atlassian

Usage

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "atlassian": {
      "command": "npx",
      "args": ["-y", "@ama-work/mcp-atlassian"],
      "env": {
        "ATLASSIAN_EMAIL": "you@company.com",
        "ATLASSIAN_API_TOKEN": "your-api-token",
        "ATLASSIAN_CLOUD_ID": "your-cloud-id"
      }
    }
  }
}

Works with bunx too — replace npx with bunx and remove -y.

Standalone mode

Provide your Atlassian credentials directly via environment variables.

1. Create an API token:

Go to Atlassian API Tokens and create a new token.

2. Find your Cloud ID:

Your Cloud ID is the first part of your Atlassian URL. For example, if your Jira is at https://mycompany.atlassian.net, your Cloud ID is mycompany.

3. Configure the env vars:

ATLASSIAN_EMAIL=you@company.com
ATLASSIAN_API_TOKEN=ATATT3xFfGF0...
ATLASSIAN_CLOUD_ID=mycompany

Registry-integrated mode

If you run ama-mcp-registry, the server fetches its token automatically:

{
  "mcpServers": {
    "atlassian": {
      "command": "npx",
      "args": ["-y", "@ama-work/mcp-atlassian"],
      "env": {
        "REGISTRY_URL": "https://your-registry.example.com",
        "INSTANCE_ID": "your-instance-id",
        "GATEWAY_PSK": "your-pre-shared-key"
      }
    }
  }
}

The token is cached and proactively refreshed 5 minutes before expiry.

MCP Registry Integration

Import the canonical definition via URL:

https://raw.githubusercontent.com/Ask-Me-Anything-dot-work/ama-mcp-atlassian/main/definitions/atlassian.json

Tools

Jira: jira_search_issues, jira_get_issue, jira_create_issue, jira_update_issue, jira_transition_issue, jira_add_comment, jira_get_projects

Confluence: confluence_search, confluence_get_page, confluence_create_page, confluence_update_page, confluence_add_comment, confluence_get_spaces

Meta: atlassian_whoami

Tool arguments pass through the MCP gateway verbatim — gateway namespacing ({instanceId}.{toolName}) does not affect argument forwarding. Past bugs where confluence_create_page dropped spaceId (400 INVALID_REQUEST_BODY) were server-side request mapping in this repo, fixed against the confluence.js v2 client contract ({ body: { spaceId, title, ... } } nesting).

Development

bash bootstrap.sh
bun run typecheck
bun run lint

License

MIT

Available Tools

14 tools
atlassian_whoamiA

Returns the identity of the currently authenticated Atlassian user — account ID, display name, email, and accessible site URLs.

Inputs: no named parameters.

confluence_add_commentB

Add a footer comment to a Confluence page.

Inputs: body (required), pageId (required).

confluence_create_pageA

Create a new Confluence page in the specified space.

Inputs: body (required), title (required), spaceId (required), status, +1 more.

confluence_get_pageA

Get the full content of a Confluence page by its numeric page ID. Returns title, spaceKey, body as markdown, version number, and last-modified date.

Inputs: pageId (required).

confluence_get_spacesA

List Confluence spaces accessible to the authenticated user. Returns space ID, key, name, type, and homepage ID.

Inputs: type, limit.

confluence_searchA

Search Confluence content using CQL (Confluence Query Language). Returns matching pages, blog posts, and spaces with title, spaceKey, URL, and a short excerpt.

Inputs: cql (required), limit.

confluence_update_pageA

Update the title and/or content of an existing Confluence page. Requires the current version number — Confluence rejects updates that omit it.

Inputs: body (required), title (required), pageId (required), versionNumber (required).

jira_add_commentB

Add a comment to a Jira issue.

Inputs: body (required), issueKey (required).

jira_create_issueA

Create a new Jira issue in the specified project.

Inputs: summary (required), issueType (required), projectKey (required), labels, +4 more.

jira_get_issueA

Get full details of a single Jira issue by its key (e.g. PROJ-123). Returns fields including summary, description, status, assignee, comments, and subtasks.

Inputs: issueKey (required), fields.

jira_get_projectsA

List all Jira projects accessible to the authenticated user. Returns project key, name, type, and lead.

Inputs: query, maxResults.

jira_search_issuesA

Search Jira issues using JQL. Returns a list of matching issues with key, summary, status, assignee, and priority.

Inputs: jql (required), fields, maxResults.

jira_transition_issueA

Move a Jira issue to a new status by triggering a workflow transition (e.g. In Progress, Done).

Inputs: issueKey (required), transitionName (required).

jira_update_issueC

Update fields on an existing Jira issue.

Inputs: issueKey (required), labels, summary, priority, +2 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A3.7/5.0

    Scored across 14 tools

    Disambiguation5/5

    Every tool is clearly scoped by domain prefix (jira_, confluence_, atlassian_) and a distinct action, so there is no real overlap. Similar tools like jira_add_comment and confluence_add_comment are separated by their domain prefix, and analogous tools like jira_get_projects vs confluence_get_spaces target different resources.

    Naming Consistency4/5

    Tool names follow a predictable domain_verb_noun pattern in snake_case, such as jira_search_issues and confluence_create_page. The only minor deviation is atlassian_whoami, which uses a command-style noun rather than a verb_noun structure, but it remains readable and clearly namespaced.

    Tool Count5/5

    14 tools is well within the ideal range and appropriately covers two product areas (Jira and Confluence) plus an identity check. Each tool serves a distinct purpose and none feel redundant or excessive for the server's broad Atlassian scope.

    Completeness4/5

    The tool surface covers core Jira and Confluence workflows: search, create, read, update, transition/comment, plus project/space discovery and identity. Minor gaps exist such as no delete operations, no listing of available Jira transitions, and no page comment retrieval, but agents can still accomplish most common tasks without dead ends.

    Maintenance

    ActivityMaintained
    ResponsivenessUsually handled

    Resources

    Related MCP Connectors

    Related MCP Servers