Claude Agentic Framework
Parallel agent swarms for Claude Code. Ship features faster with coordinated multi-agent workflows. Then start Claude Code: The fastest way to go from idea to shipped code. One agent thinks, then many agents build. Use a persona to explore the problem space and draft your planning prompt: This produces artifacts (PR-FAQs, PRDs, ADRs) that feed the next step. Launches 3-6 explorer agents in parallel. They research existing patterns, dependencies, and constraints simultaneously, then produce a ...
| Entity Passport | |
| Registry ID | gh-tool--dralgorhythm--claude-agentic-framework |
| Provider | github |
Cite this tool
Academic & Research Attribution
@misc{gh_tool__dralgorhythm__claude_agentic_framework,
author = {dralgorhythm},
title = {Claude Agentic Framework Tool},
year = {2026},
howpublished = {\url{https://github.com/dralgorhythm/claude-agentic-framework}},
note = {Accessed via Free2AITools Knowledge Fortress}
} đŦTechnical Deep Dive
Full Specifications [+]âž
Quick Commands
git clone https://github.com/dralgorhythm/claude-agentic-framework git clone https://github.com/dralgorhythm/claude-agentic-framework âī¸ Nexus Index V2.0
đŦ Index Insight
FNI V2.0 for Claude Agentic Framework: Semantic (S:50), Authority (A:0), Popularity (P:0), Recency (R:0), Quality (Q:0).
Verification Authority
đ Specs
- Language
- Shell
- License
- Open Source
- Version
- 1.0.0
Usage documentation not yet indexed for this tool.
đ View Source Code âTechnical Documentation
Claude Agentic Framework
Parallel agent swarms for Claude Code. Ship features faster with coordinated multi-agent workflows.
Install
cd your-project
curl -sSL https://raw.githubusercontent.com/dralgorhythm/claude-agentic-framework/main/scripts/init-framework.sh | bash -s .
Then start Claude Code: claude
The Swarm Workflow
The fastest way to go from idea to shipped code. One agent thinks, then many agents build.
1. Research â Single agent, deep thinking
Use a persona to explore the problem space and draft your planning prompt:
/architect redesign the auth system
This produces artifacts (PR-FAQs, PRDs, ADRs) that feed the next step.
2. Plan â Parallel exploration
/swarm-plan implement user notifications per the PRD
Launches 3-6 explorer agents in parallel. They research existing patterns, dependencies, and constraints simultaneously, then produce a decomposed plan tracked as Beads.
3. Execute â Parallel implementation
/swarm-execute
Picks up the planned work and fans out across multiple builder agents. Each worker claims a task, implements it, runs quality gates (tests, lint, types, build), and pushes to remote. Up to 8 workers in parallel.
4. Review â Adversarial, multi-pass
/swarm-review # run 2-3 times
Each pass launches 5 parallel reviewers covering different perspectives: security (OWASP Top 10), performance, architecture, test coverage, and code quality. Repeat until clean. Then ship the PR.
Full cycle
/architect â /swarm-plan â /swarm-execute â /swarm-review (2-3x) â PR
Swarm Workers
Eight specialized agent types, each tuned for cost and capability:
| Worker | Model | Use |
|---|---|---|
worker-explorer |
Haiku | Fast codebase search |
worker-builder |
Sonnet | Implementation |
worker-reviewer |
Sonnet | Code review |
worker-tester |
Sonnet | Test writing |
worker-researcher |
Sonnet | Web research |
worker-architect |
Opus | Complex design decisions |
worker-security |
Sonnet | Security analysis |
worker-refactor |
Sonnet | Code cleanup |
Haiku for read-only exploration. Sonnet for most work. Opus only for hard architectural calls.
Personas
For single-agent work, ideation, and research:
/architect /builder /qa-engineer
/security-auditor /ui-ux-designer /code-check
/swarm-plan /swarm-execute /swarm-review
Personas produce artifacts in artifacts/ that chain naturally into swarm workflows.
Also Included
- 65+ Skills â Auto-suggested based on context (designing-systems, debugging, react-patterns, ...)
- Beads â Lightweight issue tracking that coordinates swarm workers via git
- MCP Servers â Browser debugging, GitHub integration, structured reasoning, library docs
/code-checkâ Holistic codebase audit for SOLID, DRY, consistency
After Install
- Open
CLAUDE.mdand add your build/test commands - Required: Edit
.claude/rules/tech-strategy.mdto match your tech stack - Try the swarm workflow on your next feature
Docs
Social Proof
AI Summary: Based on GitHub metadata. Not a recommendation.
đĄī¸ Tool Transparency Report
Verified data manifest for traceability and transparency.
đ Identity & Source
- id
- gh-tool--dralgorhythm--claude-agentic-framework
- source
- github
- author
- dralgorhythm
- tags
- agenticagentsaiclaudedeveloper-toolsdevelopmentflowshell
âī¸ Technical Specs
- architecture
- null
- params billions
- null
- context length
- null
- pipeline tag
- other
đ Engagement & Metrics
- likes
- 0
- downloads
- 0
- github stars
- 21
Free2AITools Constitutional Data Pipeline: Curated disclosure mode active. (V15.x Standard)