waymark
Waymark is a local-first, Git-friendly work tracker that keeps project work items in Markdown files, with support for hierarchies, dependencies, and customizable fields.
107stars
75forks
Rust
Added 11/11/2026
Available after the catalog security scan passes.
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
Maintainer of this skill?
Installation
# Add to your Claude Code skills
git clone https://github.com/QiYueran-source/waymarkGetting Started
Guides for using mcp servers skills like waymark.
- Best MCP Servers in 2026Category-by-category picks: databases, dev tools, productivity, browser automation.
- What is an AI Skills Marketplace?Definitions, how marketplaces work, and how to choose between them in 2026.
- Getting Started with AI SkillsFirst-time install walkthrough for Claude Code, Codex CLI, and ChatGPT.
Compare with similar skills
See how waymark compares with popular alternatives.
Waymark
Work items are Markdown files in your repo. No database, no daemon —
git clone is the whole install.
English | 简体中文
$ waymark init
initialised Waymark project at /home/you/demo
$ waymark create --title "Ship the first screen" --category task --owner "You"
created wm-f3c4djp65z9e
$ sed -n '1,5p' work/task/wm-f3c4djp65z9e.md
---
schema: 1
id: wm-f3c4djp65z9e
title: Ship the first screen
category: task
- The documents are the source of truth. One Markdown file per work item,
YAML frontmatter, under
work/; configuration in.waymark/. Git holds the history, so a ticket can be reviewed, diffed, and reverted like code. - Reads come from an in-memory projection — and say so. Collection queries
are answered from a projection rebuilt from the files and verified against
their content fingerprints. A long-running view refreshes incrementally, so a
document whose bytes did not change is not re-parsed. Every response records
its
source,completeness, andfreshness, and a partial view fails withINCOMPLETE_PROJECTIONinstead of being published. - Agents work on the same project.
waymark-mcpis a Model Context Protocol server over stdio, so Claude Code, Cursor, or any MCP client reads and writes the work items you do. - Nothing to install at run time. Prebuilt binaries for Linux, macOS, and Windows; no database, no daemon, and no network listener except the read-only Web view, which binds loopback only.
Pre-1.0: the document format and the CLI may still change between releases. The
macOS and Windows binaries are unsigned, so verify every archive against its
.sha256 before running it.
Not for you if you need a board, notifications, or real-time multi-user collaboration — those are deliberate omissions, not a roadmap gap.
Packages
Both variants are built from the same tag and read the same project format:
- Basic — the
waymarkcommand-line tool and thewaymark-mcpModel Context Protocol server. - Web — the same commands plus the read-only Web view served by
waymark serve.
A prebuilt Waymark needs no Rust toolchain, and the Web package needs no Node.js or npm. Neither variant downloads or installs anything at run time.
Install
Prebuilt archives are attached to the GitHub Releases for
https://github.com/QiYueran-source/waymark. Download the basic archive for your
platform, verify it, extract it, and put the extracted directory on PATH.
With a Rust toolchain, the same binaries also install from crates.io:
cargo install waymark-cli # the `waymark` binary
cargo install waymark-mcp # the MCP server
The CLI crate is waymark-cli, not waymark: waymark on crates.io is an
unrelated crate.
The archive names are:
| Platform | Archive |
|---|---|
| Linux x86_64 | waymark-v<version>-linux-x86_64.tar.gz |
| macOS, Apple silicon | waymark-v<version>-macos-aarch64.tar.gz |
| macOS, Intel | waymark-v<version>-macos-x86_64.tar.gz |
| Windows x86_64 | waymark-v<version>-windows-x86_64.zip |
Each Unix archive contains no top-level directory: it holds waymark,
waymark-mcp, README.md, README.zh-CN.md, SECURITY.md, LICENSE-MIT,
LICENSE-APACHE, and THIRD_PARTY_LICENSES.txt at its root. The Windows
archive holds the same files with the binaries named waymark.exe and
waymark-mcp.exe.
THIRD_PARTY_LICENSES.txt is the dependency-licence inventory for that build.
The examples below use version 0.1.0. Set VERSION to the release you are
installing.
Linux x86_64
VERSION=0.1.0
ARCHIVE="waymark-v${VERSION}-linux-x86_64.tar.gz"
curl -LO "https://github.com/QiYueran-source/waymark/releases/download/v${VERSION}/${ARCHIVE}"
curl -LO "https://github.com/QiYueran-source/waymark/releases/download/v${VERSION}/${ARCHIVE}.sha256"
sha256sum --check "${ARCHIVE}.sha256"
mkdir -p "$HOME/.local/waymark"
tar -xzf "${ARCHIVE}" -C "$HOME/.local/waymark"
export PATH="$HOME/.local/waymark:$PATH"
sha256sum --check must print OK for the archive and nothing else. A
checksum mismatch means the download is corrupt or not the published file; do
not run it.
macOS, Apple silicon
VERSION=0.1.0
ARCHIVE="waymark-v${VERSION}-macos-aarch64.tar.gz"
curl -LO "https://github.com/QiYueran-source/waymark/releases/download/v${VERSION}/${ARCHIVE}"
curl -LO "https://github.com/QiYueran-source/waymark/releases/download/v${VERSION}/${ARCHIVE}.sha256"
shasum -a 256 -c "${ARCHIVE}.sha256"
mkdir -p "$HOME/.local/waymark"
tar -xzf "${ARCHIVE}" -C "$HOME/.local/waymark"
export PATH="$HOME/.local/waymark:$PATH"
macOS, Intel
VERSION=0.1.0
ARCHIVE="waymark-v${VERSION}-macos-x86_64.tar.gz"
curl -LO "https://github.com/QiYueran-source/waymark/releases/download/v${VERSION}/${ARCHIVE}"
curl -LO "https://github.com/QiYueran-source/waymark/releases/download/v${VERSION}/${ARCHIVE}.sha256"
shasum -a 256 -c "${ARCHIVE}.sha256"
mkdir -p "$HOME/.local/waymark"
tar -xzf "${ARCHIVE}" -C "$HOME/.local/waymark"
export PATH="$HOME/.local/waymark:$PATH"
Windows x86_64
The Windows archives are .zip files built with the MSVC toolchain. In
PowerShell:
$Version = "0.1.0"
$Archive = "waymark-v$Version-windows-x86_64.zip"
$Base = "https://github.com/QiYueran-source/waymark/releases/download/v$Version"
Invoke-WebRequest "$Base/$Archive" -OutFile $Archive
Invoke-WebRequest "$Base/$Archive.sha256" -OutFile "$Archive.sha256"
$Expected = (Get-Content "$Archive.sha256") -split '\s+' | Select-Object -First 1
$Actual = (Get-FileHash $Archive -Algorithm SHA256).Hash.ToLower()
if ($Actual -ne $Expected) { throw "checksum mismatch for $Archive" }
Expand-Archive $Archive -DestinationPath "$HOME\waymark"
$env:Path = "$HOME\waymark;$env:Path"
Get-FileHash must match the first field of the .sha256 file; a mismatch
means the download is corrupt or not the published file, so do not run it. The
$env:Path change lasts for the current session; add the directory through
System Properties → Environment Variables to keep it.
The Windows binaries are not code-signed. SmartScreen may show “Windows protected your PC” on the first launch; after the checksum matched, choose More info → Run anyway. Do not disable SmartScreen globally to run Waymark.
macOS: Gatekeeper and an unsigned binary
The macOS archives are prebuilt but not signed with an Apple Developer ID
and not notarized. macOS therefore treats waymark and waymark-mcp as
software from an unidentified developer. This is expected, not a sign that the
download is damaged.
If you download and unpack with a browser, the first launch is blocked by Gatekeeper with a message such as:
“waymark” cannot be opened because the developer cannot be verified.
or, on newer macOS versions:
Apple could not verify “waymark” is free of malware that may harm your Mac or compromise your privacy.
To allow the binary after checking that its SHA-256 matched:
- Open System Settings → Privacy & Security.
- Find the blocked
waymarkmessage and click Open Anyway. - Authenticate, then run the command again.
Gatekeeper checks each executable separately, so the first time an MCP client
launches waymark-mcp you may need to allow that binary the same way.
If the quarantine attribute is still present and you prefer the command line, remove it from the verified binaries:
xattr -d com.apple.quarantine "$HOME/.local/waymark/waymark" \
"$HOME/.local/waymark/waymark-mcp"
Only do this after the .sha256 check passed and you trust the source. Do not
disable Gatekeeper globally to run Waymark.
First commands
Check the installation, then create a project and its first work item:
waymark version
# waymark 0.1.0 (cli, mcp)
mkdir -p "$HOME/waymark-demo"
cd "$HOME/waymark-demo"
waymark init
# initialised Waymark project at /home/you/waymark-demo
# configuration /home/you/waymark-demo/.waymark/config.toml
# local settings /home/you/waymark-demo/.waymark/user.toml
# category task work/task
waymark create --title "Try Waymark" --category task --owner "Your Name"
# created wm-...
# /home/you/waymark-demo/work/task/wm-....md
waymark list
# wm-... open ready Try Waymark
waymark init creates .waymark/config.toml and the configured category
directories under work/. waymark create requires a title, a declared
category, and an owner. waymark list reads active work items.
waymark --help lists every command in this release, and
waymark agent-doc prints the same command reference as a Markdown capability
document for agents.
The Web view
The Web view is a read-only browser view of a project. It is available only in the Web package; it is not compiled into the basic package.
What a basic build does
waymark serve exists in both packages, so the command is never a surprise:
waymark serve
# error: <message naming the Web package to install>
The basic build exits with the registered WEB_NOT_INCLUDED error (validation
exit class, process status 3) and names the Web package to install. It does not
download, install, or modify anything, and it does not start a server.
Install the Web package
Download the Web archive for your platform from the same release:
waymark-web-v<version>-linux-x86_64.tar.gzwaymark-web-v<version>-macos-aarch64.tar.gzwaymark-web-v<version>-macos-x86_64.tar.gzwaymark-web-v<version>-windows-x86_64.zip
Verify its .sha256 file the same way as the basic archive, then extract it
over the directory that already holds the basic installation:
VERSION=0.1.0
WEB_ARCHIVE="waymark-web-v${VERSION}-linux-x86_64.tar.gz" # or the macOS archive
curl -LO "https://github.com/QiYueran-source/waymark/releases/download/v${VERSION}/${WEB_ARCHIVE}"
curl -LO "https://github.com/QiYueran-source/waymark/releases/download/v${VERSION}/${WEB_A
Frequently Asked Questions
What is waymark?
waymark is an open-source mcp servers skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by QiYueran-source. Waymark is a local-first, Git-friendly work tracker that keeps project work items in Markdown files, with support for hierarchies, dependencies, and customizable fields. It has 107 GitHub stars.
Is waymark safe to use?
waymark's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
How do I install waymark?
Clone the repository with "git clone https://github.com/QiYueran-source/waymark" and add it to your Claude Code skills directory (see the Installation section above).
What programming language is waymark written in?
waymark is primarily written in Rust. It is open-source under QiYueran-source on GitHub, so you can review or fork the full source.
Are there alternatives to waymark?
Yes. SkillsLLM lists many other MCP Servers skills you can browse and compare side by side. Open the MCP Servers category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh waymark against similar tools.
Comments (0)
to leave a comment.
No comments yet. Be the first to share your thoughts!
Related Skills
Reverse engineer anything with agents, from app behavior down to native binaries.
448,394
110,517
TypeScript
AI Agentsagent-skillsai
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
300,442
44,354
JavaScript
AI Agentsai-agentsanthropic
Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
209,988
61,430
TypeScript
MCP Serversaiapis
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
185,940
28,768
JavaScript
AI Agentsai-agentsanthropic
A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode,DeepSeek Harness and more. Only official website: ccswitch.io
153,243
10,069
Rust
AI Agentsai-toolsclaude-code
Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.
145,289
13,912
Python
AI Agentsai-agentsantigravity
Popular in MCP Servers
Top skills in this category by stars
🕷️ The BEST Web Scraping framework that handles everything from a single request to a full-scale crawl! Don't be shy, join here: https://discord.gg/EMgGbDceNQ and follow here for daily tips and tricks: https://x.com/Scrapling_dev
100,838
10,532
Python
MCP Serversaiai-scraping
⭐AI-driven public opinion & trend monitor with multi-platform aggregation, RSS, and smart alerts.🎯 告别信息过载，你的 AI 舆情监控助手与热点筛选工具！聚合多平台热点 + RSS 订阅，支持关键词精准筛选。AI 智能筛选新闻 + AI 翻译 + AI 分析简报直推手机，也支持接入 MCP 架构，赋能 AI 自然语言对话分析、情感洞察与趋势预测等。支持 Docker ，数据本地/云端自持。集成微信/飞书/钉钉/Telegram/邮件/ntfy/bark/slack 等渠道智能推送。
63,563
24,924
Python
MCP Serversaibark