claude-mods plugin
claude-mods adds 112 skills, 3 agents and 3 commands to Claude Code: stacks, agent operation, defense, session continuity.
- 81 ★ on the repository
- Last commit
Broad access
Why this level
- Skills run scripts, create worktrees and can start headless agents in isolated worktrees.
- Skills for autonomous loops and branch landing act on the whole repository.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/0xdarkmatter-claude-modsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Skills run scripts, create worktrees and can start headless agents in isolated worktrees.
- Skills for autonomous loops and branch landing act on the whole repository.
This is third-party code. Review the repository files before installing.
What it does
claude-mods is a Claude Code plugin built on one idea: a skill that only tells the model something stays a suggestion, while a skill with a gate gives a guarantee. It has 112 skills, three subagents, three commands, rules, hook examples and output styles. The core is agent-operation skills: a cheap headless `claude -p` in an isolated git worktree, a landing queue for parallel branches with mandatory tests, a risk ladder for autonomous loops and single-metric iteration. Another layer defends the agent: `supply-chain-defense` with a seven-day cooldown before new package versions, `prompt-injection-defense` with a hidden-Unicode scanner, and rules for pushes and worktrees. Stack skills cover React, Next.js, TypeScript, Go, Rust, Laravel, Hono, Cloudflare Workers, PostgreSQL, SQLite and Terraform with a dated fact check. For moving sessions between machines there are `/save` and `/sync`.
Who it is for. Claude Code developers who run parallel agents and want one set of checked skills for stacks and security.
Good fit when
- Several agents work in different branches and merges have to pass through gates.
- Before installing a new package a cooldown and a behavior check are needed.
- Instruction files are scanned for hidden Unicode before commit.
- A Next.js, Hono or PostgreSQL project needs a skill with current practices.
Not a fit when
- One small skill is needed: the set is big and better attached selectively.
- The agent is not Claude Code: the plugin is built for it.
- Only descriptions without scripts are wanted: the bet is on skills with code and checks.
Example request
Prepare a worktree for a parallel agent and show how to verify its branch before merging into main.Limitations
The set is big: 112 skills, but the descriptions take about 10 thousand tokens and the rest load on demand. The plugin is built for Claude Code, and in other agents skills work through the Agent Skills standard without commands and hooks. Some skills run scripts and tests worth reading before enabling. Fact marks of stack skills go stale, so 29 skills carry a freshness check. The `fleetflow` skill has moved to a separate repository. The last commit was made on 7 November 2026.
How to disable. Remove the plugin with `/plugin uninstall claude-mods@0xDarkMatter-claude-mods`. Hooks and rules added manually and the command-line tools are removed separately.
FAQ
How do I install claude-mods?
With two commands in Claude Code: `/plugin marketplace add 0xDarkMatter/claude-mods` and `/plugin install claude-mods@0xDarkMatter-claude-mods`.
How many tokens does the set take?
The descriptions of 112 skills take about 10 thousand tokens. References load only when a skill is needed.
What does fleet-ops do?
It runs a sequential landing queue for parallel branches with test checks. When the test command is not set, the landing is refused.
What does supply-chain-defense protect from?
From fresh malicious package versions: it keeps a seven-day cooldown and checks behavior, because vulnerability databases lag behind new attacks.
Which stacks are covered?
React, Vue, Next.js, TypeScript, Go, Rust, Laravel, Payload, Hono, Cloudflare Workers, PostgreSQL, SQLite and Terraform.
Discussion
No questions or reviews yet.
Related
PluginsMulti-agent orchestration for Claude Code: teams, roles, autopilot and task routing across skills and subagents
Skills, mods, subagents and hooks for Claude Code: the agent inspects your setup and installs what fits after your approval.
A Claude Code plugin marketplace with ready sets of agents, slash commands, hooks and skills, plus a search site
Workflows for Claude Code: /do, /omo, /bmad-pilot and more commands, with Codex, Gemini or OpenCode writing the code through codeagent-wrapper
A set of /sc slash commands, agents and behavior modes for Claude Code, from brainstorming and research to implementation and tests
A template for Claude Code projects: parallel agent swarms, workflow skills, safety hooks and rules in one install.
README in short
The README names four pillars of the set: running agents, defending the agent, verifiable freshness of knowledge and stack skills. Then come the repository layout, plugin installation and a list of recent releases.