Claude Agentic Framework plugin

A template for Claude Code projects: parallel agent swarms, workflow skills, safety hooks and rules in one install.

  • 130 ★ on the repository
  • Last commit
Plugin

Extended access

Why this level

  • The script copies `.claude/`, `.mcp.json`, `CLAUDE.md` and `AGENTS.md` into the project, and hooks run commands while the agent works.
  • Swarms launch up to eight parallel builders that write code and spend tokens.
All reasons and checks

dralgorhythm/claude-agentic-framework

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add plugins/claude-agentic-framework

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Extended accessThe tool runs code or makes network calls and only reads external data.

  • The script copies `.claude/`, `.mcp.json`, `CLAUDE.md` and `AGENTS.md` into the project, and hooks run commands while the agent works.
  • Swarms launch up to eight parallel builders that write code and spend tokens.

This is third-party code. Review the repository files before installing.

What it does

Claude Agentic Framework puts a ready scaffold for Claude Code into a project: skills, five worker agents, safety hooks, rules under `.claude/rules/` and a `CLAUDE.md` template. It installs once and has no dependencies. Commands split into roles and swarms. Roles such as `/architect`, `/builder`, `/qa-engineer`, `/security-auditor`, `/ui-ux-designer` and `/code-check` run a single expert. Swarms spread the work: `/swarm-plan` launches three to six explorers, `/swarm-execute` distributes tasks across builders, up to eight in parallel, `/swarm-review` seats five reviewers, and `/swarm-research` runs a multi-source investigation. All twelve workflow skills run only when the user types the command. The rules forbid piping an installer into a shell and block reading secret files.

Who it is for. Teams on Claude Code that want shared agent roles, review and safety rules from the first day of a project.

Good fit when

  • A new project starts with a ready set of roles, hooks and rules instead of manual assembly.
  • A large feature is split into small tasks and builders execute them in parallel.
  • Before merging, five reviewers check the change: security, performance, architecture, tests and quality.
  • A repository needs `.env` and other secret files closed with `permissions.deny` rules.

Not a fit when

  • The project runs on another agent and `.claude/skills/` skills are not read.
  • The team needs a licence for the template files: the repository has none.
  • Tokens for parallel agents are costly and the tasks are small enough for a single request.

Example request

Run /swarm-plan for a report-export feature, then /swarm-execute on the resulting plan.

Limitations

A full install is only done by copying files with the `init-framework.sh` script: the plugin through `/plugin install` does not add the `permissions.deny` rules or the `.claude/rules/` folder. Plugin skills carry the `/agentic-framework:` prefix, and plugin agents ignore `permissionMode` from the frontmatter. Configure `.claude/rules/tech-strategy.md` before use: the framework follows whatever you write there. The repository has no licence file, so usage rights are not defined. Swarms with eight parallel builders spend many tokens.

How to disable. For the plugin, uninstall through the `/plugin` menu. For a full install, delete the copied `.claude/`, `.mcp.json`, `CLAUDE.md` and `AGENTS.md`, or restore them from git.

FAQ

How does the full install differ from the plugin?

The plugin installs skills, five agents and hooks. The full install also adds `.claude/settings.json` with `permissions.deny` rules, the `.claude/rules/` folder, `CLAUDE.md` and `.mcp.json`.

How do I install the framework in Claude Code with a command?

Run `/plugin marketplace add dralgorhythm/claude-agentic-framework`, then `/plugin install agentic-framework@agentic-framework`. Skills arrive with the `/agentic-framework:` prefix.

Can the skills run without Claude Code?

Some clients, such as GitHub Copilot agents and VS Code, read `.claude/skills/` as is. They skip frontmatter fields specific to Claude Code.

How do I get updates?

For the plugin, run `/plugin update` after a tagged release and read `MIGRATION.md`. For a full install, run `git pull` in the clone and run the script again.

Which commands launch swarms?

`/swarm-plan` builds a plan, `/swarm-execute` carries it out with builders, `/swarm-review` seats five reviewers, and `/swarm-research` runs multi-source research.

Discussion

No questions or reviews yet.

Plugins
Editors’ pick

Multi-agent orchestration for Claude Code: teams, roles, autopilot and task routing across skills and subagents

PluginBroad access40.6KRepository stars

A set of /sc slash commands, agents and behavior modes for Claude Code, from brainstorming and research to implementation and tests

PluginExtended access23.9KRepository stars

A Claude Code workspace folder: hooks with a local Ollama detect the project type, pick skills and remind you to update the docs

WorkflowExtended access305Repository stars

Workflows for Claude Code: /do, /omo, /bmad-pilot and more commands, with Codex, Gemini or OpenCode writing the code through codeagent-wrapper

PluginBroad access2.8KRepository stars

A meta-skill for Claude Code that turns a domain description into an agent team and the skills it uses, using six team-architecture patterns

SkillMinimal access9.3KRepository stars
Editors’ pick

Anthropic's coding agent for the terminal, IDE and GitHub: understands your codebase, runs tasks and handles git

CLIBroad access154.7KRepository stars

Comparison with similar tools

Cards that share the main search query, and how they differ.

ToolOfficialRepository starsMain difference
Claude Agentic Framework (this card)no130
SuperClaude Frameworkno23,928a set of /sc slash commands and modes for Claude Code; here the focus is agent swarms, safety hooks and permission rules
oh-my-claudecodeno40,625orchestration of agent teams with an autopilot; here a project template with roles, review by five reviewers and deny rules

README in short

The design leans on the DORA 2025 report: AI amplifies the engineering discipline that already exists, while large unreviewed changes erode stability. Quality gates are therefore mandatory, and work is split into small tasks checked from several angles.

README badge

Report a problem

Foxx AIClaude Agentic Framework
Foxx AI

I am Foxx AI and I have already vetted this tool. Ask about install, setup or anything else, and I will keep it simple.