agent-system-design
Skill1 starsChanged 2 months agoMITView on GitHub
Use when designing, architecting, or benchmarking multi-tier autonomous AI Agent systems with layered memory, semantic caching, model routing, and deterministic guardrails.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- hemapriyan-rk/claude-rigor-skillstheir own version, first seen Sep 26, 2026
- antonioneto11/agent-dev-frameworktheir own version, first seen Jul 11, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- hemapriyan-rk/claude-rigor-skills★ 1same name
- antonioneto11/agent-dev-framework★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository11 other files it gives its agents.
- agent-reachSkill
- cloudflare-fullstack-opsSkill
- cloudflare-worker-api-proxySkill
- humanizerSkill
- human-voice-writingSkill
- large-file-data-analysisSkill
- markdown-to-html-reportSkill
- multi-platform-distributionSkill
- telegram-pipe-table-renderingSkill
- tiered-memorySkill
- vps-bootstrapSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 61
Unbounded Agent Loops: Always enforce max_turns (e.g.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Autonomous Agent System Architecture Design
Overview
A battle-tested production blueprint for engineering robust, resilient AI coding agents and autonomous workflows. It provides architectural patterns across 4-layer memory hierarchy, semantic caching, cost/token budgeting, dynamic model routing, and lifecycle guardrails.
When to Use & When NOT to Use
When to Use
- Building new autonomous coding agents or multi-agent orchestration frameworks.
- Diagnosing agent infinite loops, context overflow, memory drift, or high API costs.
- Implementing graceful fallbacks across diverse LLM providers.
When NOT to Use
- Single-turn prompt engineering with no persistent toolchain or session state.
- Simple script automation that does not require an LLM reasoning loop.
Core Architectural Pillars
┌─────────────────────────────────────────────────────────────┐
│ User Interface / Gateway │
└──────────────────────────────┬──────────────────────────────┘
│
┌──────────────────────────────▼──────────────────────────────┐
│ Execution Kernel / Loop │
│ ┌───────────────────────┐ ┌───────────────────────┐ │
│ │ Model Routing & Fallback │◄────►│ Lifecycle Guardrails │ │
│ └───────────────────────┘ └───────────────────────┘ │
│ ┌───────────────────────┐ ┌───────────────────────┐ │
│ │ Tiered Memory Engine │◄────►│ Semantic / Tool Cache │ │
│ └───────────────────────┘ └───────────────────────┘ │
└──────────────────────────────┬──────────────────────────────┘
│
┌──────────────────────────────▼──────────────────────────────┐
│ Tools & Isolated Sandbox │
└─────────────────────────────────────────────────────────────┘| Architecture Layer | Core Responsibility | Failure Mode if Missing |
| Model Routing & Fallback | Multi-tier failover (Primary -> Fallback 1..N) with automatic health probes | Total outage when single upstream provider 429s/500s |
| Tiered Memory Engine | Separates L1 Core facts from L2/L3 on-demand notes & session history | Context saturation and prompt token runaway |
| Lifecycle Guardrails | Intercepts destructive system commands and prevents runaway execution | Accidental crashes or runaway subagents |
| Semantic & Tool Cache | Caches idempotent tool calls (search, web fetch, AST parsing) | Redundant roundtrips and high inference latency |
Common Pitfalls
- Unbounded Agent Loops: Always enforce
max_turns(e.g. 50-90 turns) and budget watchdog alarms. - Synchronous Tool Bottlenecks: Batch independent reads, searches, and probes into parallel tool turns.
- Hardcoding Upstream Models: Decouple model aliases from underlying provider configs to enable zero-downtime model switching.
Verification Checklist
- Multi-tier fallback chain is configured and verified via live health check probes.
- Tool execution runs within an isolated environment or has safety guardrails enabled.
- Memory layer keeps turn prompt injection under 15% of context window.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.