World Monitor
Real-time global intelligence on conflict, markets, energy, trade, shipping and cyber. 100 tools.
Unofficial MCP server for Leantime: 42 agent-first tools, backup/restore, dry-run validation
列表按使用分(满分 100)排序,由以下部分组成
| 热度 | 0/80 | GitHub 星标或每周下载量:较大的计 60,较小的计 20 |
|---|---|---|
| 新鲜度 | 7/12 | 30 天内有发布或代码更新得满分 |
| 仍在响应 | 0/8 | 上次检查时云端端点有响应 |
| 使用分 | 7/100 |
质量: 基础检查通过 3/3 项(仅展示,不用于排序) · 上升势头: 暂无数据 (只用于“上升中”标签)
选择你使用的 AI 应用,然后复制设置。
通过 3/3 项检查 · 在你的电脑上运行
第一步:你用哪个 AI?
选择你用的应用,步骤会显示在下方。
Claude 暂时无法连接 Leantime MCP (Unofficial)
开发者尚未在 MCP 注册表中填写安装方式
ChatGPT 暂时无法连接 Leantime MCP (Unofficial)
开发者尚未在 MCP 注册表中填写安装方式
B · 其他 类中能连接 ChatGPT 的其他服务器
从同类中使用较多的里挑选,仅限用地址即可连接、无需密钥的
Claude Code 暂时无法连接 Leantime MCP (Unofficial)
开发者尚未在 MCP 注册表中填写安装方式
Cursor 暂时无法连接 Leantime MCP (Unofficial)
开发者尚未在 MCP 注册表中填写安装方式
VS Code 暂时无法连接 Leantime MCP (Unofficial)
开发者尚未在 MCP 注册表中填写安装方式
依据这个服务器的连接方式和各应用自己的文档,检查于 2026年11月10日。
在你的电脑上运行,适用于
GooseLM StudioMsty StudioCherry StudioVisual StudioWindsurf (Devin Desktop)ZedKiroClineContinueJetBrains AI AssistantCodex CLIGemini CLIOpenCodenodecodeWarpLibreChatRaycast
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月19日
注册名称已通过 GitHub 账户 4lador 验证,该账户拥有此仓库
代码最后更新于2个月前
使用 MIT
它在你的电脑上运行,没有互联网端点
在你的电脑上运行,改用本地设置
它在你的电脑上运行;我们只能读取互联网端点的工具列表
如果你是作者,可把这段代码贴到 README。徽章会链接回本页。
[](https://mcp.popnx.com/en/servers/4lador-leantime-mcp)提问、分享技巧或报告无法使用。按“有帮助”数量排序。
还没有评价
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 3 项适用的基础检查中通过了 3 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
GitHub 仓库最后更新于2个月前,最新版本(2.9.0)于2个月前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Rust。
同一分类中有:World Monitor, openstatus, Oh My Posh Validator, Strata, Jitsu
同一分类、其他作者、命令已通过检查。官方的排在前面,再按星标排序。
如何看卡片:↓ 每周下载量i★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
Real-time global intelligence on conflict, markets, energy, trade, shipping and cyber. 100 tools.
Manage monitors, status pages, incidents and maintenances in your openstatus workspace.
Validate oh-my-posh configurations and segment snippets against the official schema.
MCP server for progressive tool usage at any scale (see https://klavis.ai)
Manage Jitsu data pipelines: destinations, streams, connections, functions, live events.
| 本项 Leantime MCP (Unofficial) | World Monitor | openstatus | Oh My Posh Validator | Strata | Jitsu | |
|---|---|---|---|---|---|---|
| 运行位置 | 需要安装在电脑上 | 在云端运行,无需安装 | 在云端运行,无需安装 | 在云端运行,无需安装 | 在云端运行,无需安装 | 在云端运行,无需安装 |
| 需要密钥 | 否 | 否 | 否 | 否 | 否 | 否 |
| 可在 Claude 中按 + 添加 | 否 | 否 | 否 | 否 | 否 | 否 |
| 基础检查 | 3/3 | 4/4 | 4/4 | 4/4 | 1/4 | 4/4 |
| 许可证 | MIT | AGPL-3.0 | AGPL-3.0 | MIT | Apache-2.0 | MIT |
| 价格 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 |
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
mcp-name: io.github.4lador/leantime-mcpAn unofficial Model Context Protocol server for Leantime, designed agent-first: it lets LLM tools (opencode, Claude Code, Claude Desktop, Cursor, Codex, or any MCP client) work on your Leantime projects through 42 tools with dry-run validation, bulk operations, backup/restore and one-call project context.
It is an independent, open-source integration — stdio only, running locally with your own API key and your own instance. It does not aim to replace Leantime's official offerings; if you want vendor-supported integrations, use those.
~3 MB self-contained binary (rustls — no system OpenSSL), ~1.2 ms median startup and ~5 MB idle memory (VmRSS after handshake — measured on a Linux x86_64 dev machine, n=100 spawns).
Documentation: Quick start · Configuration · Safety · Tools · Troubleshooting · Harness setup · Backup & recovery · Migration from v1.x · CHANGELOG · CONTRIBUTING · SECURITY
An example session (abridged):
You: What's the status of the Acme launch project?
Agent calls
leantime_project_context— one call returning progress, health, sprint and milestones:
`json
{ "project": { "name": "Acme launch", "progress": { "percentDone": 64.5, "ticketsTotal": 279, "ticketsDone": 180 } },
"health": { "blocked": 0, "overdue": 7, "unassigned": 3, "openTotal": 99 },
"currentSprint": { "name": "Sprint 4", "status": "current", "daysRemaining": 6, "openTickets": 8 } }
`
Agent: Acme launch is at 64.5% (180 of 279 tickets done). 99 tickets are open — 7 overdue, 3 unassigned, none blocked. Sprint 4 ends in 6 days with 8 open tickets; the milestone in progress (PHASE 8) sits at 47.8%.
leantime_project_context — the whole project picture in one call (progress, health, sprint, milestones), under 4 KBdryRun: true on every mutation — same validations, from → to diffs on updates, no write requestsidempotencyKey on creation tools — a retried call with an already-succeeded key replays the original result instead of writing a duplicatetruncated on capped lists, changed/unchanged diffs on updates — agents see what a call actually didLEANTIME_MCP_PROFILE=readonly — a server-wide execution profile that removes write capabilities entirelyLEANTIME_MCP_BACKUP_RETENTION_DAYS), atomic backup writes, and a restore manifest mapping every old→new idLEANTIME_MCP_DESTRUCTIVE_POLICY=deny as an emergency stopLinux / macOS (x86_64, aarch64):
curl -fsSL https://raw.githubusercontent.com/4lador/leantime-mcp/main/install.sh | sh
Windows (PowerShell):
irm https://raw.githubusercontent.com/4lador/leantime-mcp/main/install.ps1 | iex
Both installers verify the published SHA-256 checksum before installing and abort on mismatch. Release binaries cover all 5 targets — Linux (x86_64, aarch64), Windows x86_64 and macOS (Intel, ARM) — on the releases page.
Via cargo (requires the Rust toolchain):
cargo install leantime-mcp # installs the leantmcp binary to ~/.cargo/bin
From source:
cargo build --release # → target/release/leantmcp
leantmcp url set https://your-instance.leantime.io # once
leantmcp key set # once — hidden prompt leantmcp setup # writes the config leantmcp doctor # verify everything end-to-end
Your Leantime API key: My Account → API Keys → Generate on your instance.
| Harness | Default scope | Config file |
|---|---|---|
| opencode | global | ~/.opencode/opencode.json |
| Claude Code | project | ./.mcp.json |
| Claude Desktop | global | claude_desktop_config.json (path per OS) |
| Cursor | global | ~/.cursor/mcp.json |
| Codex | global | ~/.codex/config.toml |
Every generated config is a bare command with no credentials — the binary resolves them from the keyring at startup. Project-scoped setups, --scope/--instance/--name options and multiple instances: see Harness setup.
Credentials live in named instance profiles — ~/.config/leantime/instances/ / (api-key, mode 0600, and instance-url), with a default file naming the default. The key is read from the keyring file or environment — it is not accepted as a command-line argument (shell history) and is not logged; key commands are CLI-only and are not exposed as MCP tools.
leantmcp url show # resolved URL + where it comes from
leantmcp key show # masked display (lt_h13…Fc3O) leantmcp key test # live validation against the instance leantmcp key rotate # mint a new key (same role), verify it live, replace the stored one leantmcp instance add … # multiple Leantime instances (see Harness setup)
LEANTIME_URL / LEANTIME_API_KEY environment variables remain available as per-run overrides; LEANTIME_INSTANCE selects a profile. Resolution order: env (explicit override) → LEANTIME_INSTANCE profile → the default file.
This software is provided without warranty (MIT). It drives Leantime with your API key on your behalf — back up your Leantime data before letting agents operate on it.
confirm: true required by default; LEANTIME_MCP_DESTRUCTIVE_POLICY=deny refuses them outright, allow skips the gate for CI). Project hiding/deletion is intentionally not exposed.dryRun: true — validations run, updates resolve from → to values (with status labels), bulk tools return per-item previews, and no write request is sent. Agent guidance is built into the tool descriptions on a three-tier policy: direct execution for explicit values, dry-run-then-confirm when the agent chose the values itself, dry-run mandated for bulk.editorId validated against the real user list, or an explicit unassigned: true).tools/list (no context-window cost) and calling one is refused with an actionable error. Read-only mode: leantmcp tools disable all && leantmcp tools enable readonly. State is stored per instance profile.leantime_backup_project (MCP) or leantmcp backup (CLI) snapshot a project cheaply — agents are instructed to use it before bulk modifications. See Backup & recovery.42 tools across 9 domains: projects & clients (8), tickets (10), comments (4), time tracking (4), milestones (6), sprints (4), users (1), bulk operations (3), backup & context (2). Updates use the patch API — only provided fields change. Descriptions follow a Markdown subset converted server-side; raw HTML in input is escaped.
Supported Markdown subset (for descriptions and comments)
| Markdown | Result |
|---|---|
# … #### | Headings (levels 5+ clamp to h4) |
| blank-line separated text | Paragraphs (single newlines become line breaks) |
- item / 1. item | Unordered / ordered lists (nesting supported) |
- [ ] x / - [x] x | Interactive checklists |
**bold**, *italic*, ~~strike~~ | Emphasis |
`code` and fenced ` blocks | Inline and block code |
[label](https://…) | Links (http/https/mailto only) |
> quote | Blockquote |
--- | Horizontal rule |
Raw HTML in descriptions is escaped before reaching Leantime — it renders as literal text.
All 42 tools (click to expand)
Projects & clients
| Tool | Description |
|---|---|
leantime_list_projects | List all projects |
leantime_get_project | Get project details |
leantime_get_project_progress | Get project progress metrics |
leantime_create_project | Create a project (Markdown details, clientId required) |
leantime_update_project | Update a project (patch — only provided fields change) |
leantime_find_projects | Search projects by name |
leantime_list_project_users | List users assigned to a project (valid editorId candidates) |
leantime_list_clients | List clients (clientId needed to create projects) |
Tickets
| Tool | Description |
|---|---|
leantime_list_tickets | List tickets with filters (status, milestone, sprint, user, type, search) |
leantime_get_ticket | Get ticket details |
leantime_create_ticket | Create a ticket (Markdown description, mandatory assignment, subtasks via dependingTicketId; dryRun supported) |
leantime_update_ticket | Update a ticket (patch — only provided fields change; dryRun supported) |
leantime_delete_ticket | Delete a ticket (confirm-gated) |
leantime_list_subtasks | List a ticket's subtasks |
leantime_my_tasks | Open tickets assigned to a user (default: the API key owner) |
leantime_get_ticket_options | Priorities, efforts, kanban columns and ticket types |
leantime_get_statuses | Get status labels for a project |
leantime_get_ticket_types | Get ticket types for a project |
Comments
| Tool | Description |
|---|---|
leantime_list_comments | List a ticket's discussion |
leantime_add_comment | Comment on a ticket (Markdown converted to rich HTML) |
leantime_update_comment | Edit a comment (Markdown) |
leantime_delete_comment | Delete a comment (confirm-gated) |
Time tracking
| Tool | Description |
|---|---|
leantime_log_time | Log hours on a ticket (add accumulates, set is idempotent; dryRun supported) |
leantime_get_ticket_time | Total and per-day booked time for a ticket |
leantime_list_timesheets | List time entries between two dates |
leantime_delete_timesheet_entry | Delete a time entry (confirm-gated) |
Milestones
| Tool | Description |
|---|---|
leantime_list_milestones | List milestones of a project |
leantime_get_milestone | Get milestone details |
leantime_create_milestone | Create a milestone (Markdown description, mandatory assignment; dryRun supported) |
leantime_update_milestone | Update a milestone (patch; dryRun supported) |
| `leantime_get_milestone_progres