guardian-agent-prompts
agent
Warn
Health Warn
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 5 GitHub stars
Code Pass
- Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Pass
- Permissions — No dangerous permissions requested
Purpose
This repository provides 49 production-tested AI agent system prompts and 7 n8n workflow templates designed to help developers build and manage multi-agent orchestration systems. It serves as a prompt engineering resource rather than a standalone executable application.
Security Assessment
Overall Risk: Low. The code scan analyzed 12 files and found no dangerous patterns, hardcoded secrets, or requests for dangerous permissions. Because this tool primarily consists of text-based prompt templates and Markdown files rather than compiled software, the inherent risk of executing malicious shell commands or silently accessing local sensitive data is minimal. The user retains full control over how the AI agents interpret and execute the provided prompts.
Quality Assessment
The project is highly transparent and actively maintained, with its most recent push occurring today. It is released under the permissive and standard MIT license, making it safe for commercial and personal use. However, it currently suffers from low direct community visibility, having only accumulated 5 stars on its own repository. Despite this low direct following, the project boasts strong indirect credibility. It claims to be tested across 10,000+ tasks and has been featured and merged into massive, highly reputable community repositories with over 1 million combined stars, significantly boosting its trust level.
Verdict
Safe to use.
This repository provides 49 production-tested AI agent system prompts and 7 n8n workflow templates designed to help developers build and manage multi-agent orchestration systems. It serves as a prompt engineering resource rather than a standalone executable application.
Security Assessment
Overall Risk: Low. The code scan analyzed 12 files and found no dangerous patterns, hardcoded secrets, or requests for dangerous permissions. Because this tool primarily consists of text-based prompt templates and Markdown files rather than compiled software, the inherent risk of executing malicious shell commands or silently accessing local sensitive data is minimal. The user retains full control over how the AI agents interpret and execute the provided prompts.
Quality Assessment
The project is highly transparent and actively maintained, with its most recent push occurring today. It is released under the permissive and standard MIT license, making it safe for commercial and personal use. However, it currently suffers from low direct community visibility, having only accumulated 5 stars on its own repository. Despite this low direct following, the project boasts strong indirect credibility. It claims to be tested across 10,000+ tasks and has been featured and merged into massive, highly reputable community repositories with over 1 million combined stars, significantly boosting its trust level.
Verdict
Safe to use.
49 production-tested AI agent system prompts + 7 free n8n workflow templates. Built from 6 months running a 57-agent system 24/7. Free orchestrator included.
README.md
Guardian Agent Prompts
49 production-tested AI agent system prompts for building multi-agent orchestration systems with Claude Code.
Featured in f/prompts.chat, affaan-m/everything-claude-code, x1xhlol/system-prompts-and-models-of-ai-tools, Shubhamsaboo/awesome-llm-apps, and 247 more repositories (1M+ combined stars).
What's Inside
- 49 specialized agent system prompts covering orchestration, coding, security, trading, OSINT, business automation, VRChat integration, and more
- Anti-duplication task registry with 55% similarity threshold (SQLite-backed)
- 5-step quality gates (file diff, tests, secrets scan, build, scope check)
- 30-minute heartbeat monitoring to catch stale assignments
- Production-tested across 10,000+ tasks
Quick Start
# Clone the repository
git clone https://github.com/milkomida77/guardian-agent-prompts.git
# Browse agent prompts
ls agents/
# Use the orchestrator
cp agents/guardian-orchestrateur.md ~/.claude/agents/
Agent Catalog
| Category | Agents | Examples |
|---|---|---|
| Orchestration | 3 | orchestrateur, brain, setup-master |
| Code & Architecture | 5 | codex, architect, codex-auditor, codex-cleaner, mcp-builder |
| Security | 3 | hak, security-scan, prompt-auditor |
| Trading & Finance | 3 | trading, trading-live, market-analyzer |
| Business & Revenue | 5 | business, money, arena, arena-alpha/beta/gamma |
| Communication | 6 | discord, telegram, voice, voicelab, translator, charme |
| Infrastructure | 6 | pulse, watchdog, cloudflare, n8n, nexus, updater |
| Research | 4 | ghost, deep-research, youtube, sentinel |
| VRChat | 7 | elliott, contacts, echo, worlds, solo-leveling, vrct, avatar |
| Specialty | 7 | swim, secretaire, aquatic, inmo, comet, memory, sigma |
Featured In
Key Patterns
Anti-Duplication Registry
# SQLite-based task registry
# 55% threshold = conflict (already assigned)
# 45% threshold = warning (needs human review)
python task_registry.py check "implement auth module"
python task_registry.py claim "implement auth module" --agent guardian-codex
Quality Gate
Before marking any task "done":
git diffshows actual modifications- Test suite passes
- No API keys or credentials in diff
- Project builds and runs
- Changes stay within assigned scope
Delegation Format
[ORCHESTRATOR -> agent-name] TASK: description
SCOPE: files/directories allowed
VERIFICATION: command to prove completion
DEADLINE: timeframe
Get the Full System
The complete Guardian system with all 49 agents, task registry, quality gates, and orchestration framework:
Get it on Gumroad ($29 - Use code LAUNCH49 for $19)
Documentation
| Guide | Description |
|---|---|
| Getting Started | Setup and first agent |
| Production Patterns | All 49 agent catalog |
| Quality Gates | Verification system |
| Framework Comparison | vs CrewAI, LangGraph, AutoGen |
| n8n Integration | Workflow automation |
| Common Mistakes | 7 costly anti-patterns |
License
MIT License - see LICENSE for details.
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found