Jira Delivery

io.github.teransarathchandra由 GitHub 帳號 teransarathchandra 發布v2.0.0更新於 Oct 25, 2026

Jira Cloud delivery layer: ticket briefs, acceptance-criteria checks before the PR, QA handoff.

概覽

AI 產生的概覽

在本機執行的 Jira Cloud 交付助手:把票據轉成實作簡報,用驗收標準檢查本機變更,並產生 QA 交接與 PR 前檢查。

可以做什麼
搜尋並讀取 Jira 票據,包含父項、史詩與關聯票據的完整脈絡,並尋找相關的 Confluence 頁面與頁面摘要。把票據整理成可直接實作的簡報,包含目標、驗收標準、限制與實作提示詞。將本機程式碼差異與驗收標準及完成定義比對,並產生需求追蹤矩陣,把每條標準對應到變更檔案。產生測試策略、審查報告、QA 交接、版本說明與任務報告。辨識並儲存儲存庫慣例,將可重用的代理提示詞或指令寫入儲存庫。
適用情境
適合在 Jira Cloud 上搭配編碼代理處理票據時使用,用於結構化開工、送出 PR 前依驗收標準檢查就緒度,以及產生交接或版本產物。
執行需求
透過 npx 以 stdio 在本機執行,需要 Node.js 22 以上。需要設定 JIRA_BASE_URL、JIRA_EMAIL 與 JIRA_API_TOKEN,並能連線你的 Atlassian 網站。僅在 JIRA_ENABLE_WRITES=true 時註冊寫入工具;儲存專案模式需要 DELIVERY_PATTERN_MEMORY_ENABLED=true。
安裝前請注意
API 權杖屬於憑證,應保存在環境變數或安全儲存中。寫入工具(留言、狀態轉換、指派、關聯)預設關閉;啟用後可修改 Jira 票據,不過 README 稱寫入會先預覽並需核准。delivery_generate_agent_pack 與模式儲存會將檔案寫入你的儲存庫。票據文字被視為不可信資料。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Jira Delivery,將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Desktop only,透過 STDIO。 STDIO 服務會啟動本機處理程序,因此需要 SourceWeft 桌面主機。

其他 MCP 客戶端

參照 儲存庫 中的啟動說明。

README

[Jira Delivery MCP: from Jira ticket to merged PR without leaving your coding agent]

[npm version] [CI] [Tool-selection evals: 36/36] [Node.js 22 or newer] [MIT license]

An MCP server and agent skill for Jira Cloud. Start a ticket with a ready-to-build brief,
check your branch against its acceptance criteria before the PR, and hand off — from your coding agent.

[Install in Claude Code] [Add to Cursor] [Install in VS Code]

How it works

1 · Start a ticket

"I'm starting ABC-1, get me ready."

Readiness, blockers, acceptance criteria, next steps and a branch name — before you write a line.

[The agent lists open issues, then gets a ready-to-build brief for ABC-1]

2 · Verify before the PR

"Check my branch against ABC-1 before I open the PR."

Your local diff against the acceptance criteria and Definition of Done, with one verdict — ready_for_pr, needs_work or insufficient_evidence — and the files behind each criterion.

[The branch covers one of two criteria, so the verdict is needs_work]

3 · Hand off

"Post the QA handoff to ABC-1."

QA handoffs, release notes, test plans and reviews. With write-back on, the agent shows you a preview and waits for your approval.

[A comment is previewed; nothing changes in Jira until the user approves]

Install

You need Node.js 22+, your Jira Cloud URL, your Atlassian email and an API token.

[圖片] Claude Code — plugin with the server and the skill
/plugin marketplace add teransarathchandra/jira-mcp-server/plugin install jira-delivery@jira-delivery

It asks for your site, email and token (the token goes to secure storage). Without the plugin:

bash
claude mcp add jira-delivery \  -e JIRA_BASE_URL=https://your-domain.atlassian.net \  -e [email protected] \  -e JIRA_API_TOKEN=your-api-token \  -- npx -y jira-delivery-mcp
[圖片] Cursor

Click Add to Cursor and replace the three placeholder values, or add the JSON below to ~/.cursor/mcp.json.

[圖片] VS Code (GitHub Copilot agent mode)

Click Install in VS Code (or Insiders). VS Code asks for your site, email and token and stores the token as a secret.

Manually: add to .vscode/mcp.json — VS Code uses servers, not mcpServers:

json
{  "servers": {    "jira-delivery": {      "type": "stdio",      "command": "npx",      "args": ["-y", "jira-delivery-mcp"],      "env": { "JIRA_BASE_URL": "https://your-domain.atlassian.net", "JIRA_EMAIL": "[email protected]", "JIRA_API_TOKEN": "your-api-token" }    }  }}
[圖片] Windsurf · [圖片] Claude Desktop · any MCP client

Add to the client's MCP config — Windsurf: Settings → MCP Servers; Claude Desktop: claude_desktop_config.json:

json
{  "mcpServers": {    "jira-delivery": {      "command": "npx",      "args": ["-y", "jira-delivery-mcp"],      "env": { "JIRA_BASE_URL": "https://your-domain.atlassian.net", "JIRA_EMAIL": "[email protected]", "JIRA_API_TOKEN": "your-api-token" }    }  }}
⌨️ Codex CLI

In ~/.codex/config.toml:

toml
[mcp_servers.jira-delivery]command = "npx"args = ["-y", "jira-delivery-mcp"]env = { JIRA_BASE_URL = "https://your-domain.atlassian.net", JIRA_EMAIL = "[email protected]", JIRA_API_TOKEN = "your-api-token" }

Check your setup: with the three variables exported, npx jira-delivery-mcp doctor verifies the login and prints the custom-field ids your site needs (acceptance criteria, story points, epic). Other agents can add the jira-delivery skill, also attached to each release.

Tools

StepTools
Startjira_search_issues · jira_get_issue · jira_prepare_work_brief
Verifydelivery_check_changes · delivery_traceability_matrix
Hand offdelivery_generate · jira_add_comment* · jira_update_issue*
Contextconfluence_search · confluence_get_page · delivery_project_patterns · delivery_generate_agent_pack

* Opt-in with JIRA_ENABLE_WRITES=true. Every tool returns typed structuredContent; see the tool reference.

Safe by default

  • 🔒 Local and private — runs on your machine over stdio; talks only to your Atlassian site; no telemetry.
  • ✋ Read-only unless you opt in — and then every write is previewed, waits for your approval, is sent once and audited.
  • 🧱 Ticket text is untrusted — fenced as data, scanned for prompt injection, never turned into instructions.
  • 🗝️ Credentials stay in your environment — never logged; redacted from all output.

Documentation

Usage · Tools · Configuration · Security · Troubleshooting · Migrating from 1.x · Changelog

Contributing

Issues and pull requests are welcome — see CONTRIBUTING.md. Security reports: SECURITY.md.

MIT © Teran Sarathchandra

來源:README.md,提交 521a77b

工具

10

jira_get_issue

Get a Jira issue

Fetch a Jira issue as a developer brief: description, acceptance criteria, comments and attachments. Use depth "full" for parent/epic/linked-issue context and includeConfluence for related documentation. Use this to read a ticket. To start work, check readiness, verify changes or write a report, call jira_prepare_work_brief or the delivery_* tool directly: they fetch the issue themselves.

輸入參數depthissueKey必填includeConfluence

jira_search_issues

Search Jira issues

List open Jira issues assigned to the authenticated user, optionally limited to one project. Use this to find what to work on next. Pass jql to search with your own filter instead (e.g. "type = Bug AND sprint in openSprints() ORDER BY priority DESC"); it is limited to projectKey when given, and to the allowed projects when the strict allowlist is on. When more results exist, data.nextPageToken is set: pass it back as pageToken for the next page.

輸入參數jqlpageTokenmaxResultsprojectKey

jira_prepare_work_brief

Prepare a work brief for a Jira issue

Turn a Jira issue into a ready-to-implement brief for a coding agent: goal, acceptance criteria, constraints and an implementation prompt. Use this when starting work on a ticket, or to check whether it is ready to start (readiness, blockers, missing acceptance criteria). Fetches the issue itself.

輸入參數depthissueKey必填allowElicitationincludeConfluence

confluence_get_page

Get a Confluence page

Read a Confluence page by ID and return a cleaned, size-bounded summary of its content.

輸入參數pageId必填maxChars

delivery_check_changes

Check local changes against a Jira issue

Check work on a Jira issue: does the branch cover the acceptance criteria and meet the Definition of Done? alignment and dod compare the local diff with them — run them before opening a PR. impact predicts the affected areas and works before any code is written, so use checks ["impact"] when planning. Each check runs independently. Returns a verdict (ready_for_pr, needs_work, insufficient_evidence; not_applicable when only impact ran) and per-criterion coverage. Fetches the issue itself.

輸入參數checksissueKey必填repoPathbaseBranchcompareRefmaxDiffCharsincludeConfluence

delivery_traceability_matrix

Build a requirement traceability matrix

Build a requirements traceability matrix: map each requirement and acceptance criterion of a Jira issue to the changed files that implement it, and flag gaps. Fetches the issue itself. For a ready-for-PR verdict use delivery_check_changes.

輸入參數issueKey必填repoPathbaseBranchcompareRefincludePrDiffincludeConfluence

delivery_generate

Generate a delivery document for a Jira issue

Generate a delivery artifact from a Jira issue and the local diff: test_strategy, reviewer_report (needs persona), qa_handoff, release_notes (audience) or task_report (combined sections). Use test_strategy to plan tests and reviewer_report to review a ticket from a persona's point of view. Fetches the issue itself. Returns Markdown; writes no files.

輸入參數type必填personaaudienceissueKey必填repoPathsectionsbaseBranchcompareRefincludePrDiffincludeConfluence

delivery_generate_agent_pack

Write an agent workflow pack into the repo

Write reusable Jira delivery prompts/commands for a coding agent into the repository (claude: .claude/commands and .claude/skills, codex: .codex-prompts/, generic: .mcp-prompts/).

輸入參數client必填repoPathoverwrite

delivery_project_patterns

Scan, read or clear stored project patterns

Detect the repository’s conventions (framework, test layout, tooling) and store them so other delivery tools can tailor their output. scan = detect and store, get = read stored patterns, clear = delete them. Storing requires DELIVERY_PATTERN_MEMORY_ENABLED=true in the server environment.

輸入參數action必填repoPath

版本歷史

1
  1. v2.0.0最新Oct 25, 2026

評分

已滿足適用檢查項的 90%

該評分衡量此收錄條目的完整程度,依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價,也不表示伺服器安全或目前可用:可連通性是帶日期的檢查結果,評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通25 分
  • 已宣告認證方式10 分

    請在註冊中心條目中宣告憑證請求標頭,或保持端點能夠回應探測,以便判定其認證方式。

  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分