mirror of
https://github.com/x1xhlol/system-prompts-and-models-of-ai-tools.git
synced 2026-06-18 07:19:35 +00:00
| .. | ||
| orchestrator-system-prompt.txt | ||
| README.md | ||
Guardian AI
Type: Multi-agent orchestration system (open source) Model: Claude (Opus/Sonnet), also works with GPT, Gemini, Llama, Mistral Agents: 57 specialized agents coordinated by a single orchestrator Production: 10,000+ tasks over 6+ months Source: https://github.com/milkomida77/guardian-agent-prompts
Prompts in this directory
| File | Description |
|---|---|
| orchestrator-system-prompt.txt | The main orchestrator that coordinates 57 specialized agents (10 representative categories shown in routing table). Handles task decomposition, anti-duplication, quality gates, and parallel agent execution. |
Architecture
Guardian uses a hub-and-spoke model:
- 1 Orchestrator (this prompt) routes ALL incoming tasks
- 57 Specialized Agents handle specific domains (code, security, trading, OSINT, business, etc.)
- Task Registry prevents duplicate work across agents
- Quality Gates require verification evidence before marking tasks done
- Knowledge Graph provides persistent memory across sessions
Key Patterns
- Identity + NOT-block: Each agent defines what it IS and what it IS NOT (reduces task drift ~35%)
- Task Registry: SQLite-based anti-duplication with similarity matching
- Setup Master: Every task gets a blueprint before delegation (agents, tools, risks, order)
- Quality Gate: Agent output is a CLAIM; test output is EVIDENCE
- 30-minute Heartbeat: Orchestrator checks progress and reassigns stale tasks
Differences from other agent systems
| Feature | Guardian | Typical Agent Frameworks |
|---|---|---|
| Prompt length | 200-800 lines | 20-50 lines |
| Constraint ratio | 20-30% | <5% |
| NOT-blocks | Every agent | Rare |
| Task registry | Built-in | Not included |
| Quality gates | Mandatory | Optional |
| Error handling | Explicit per failure mode | Generic retry |