Building Agent Systems

Trust Score 78/100

Reference guide for designing single- and multi-agent LLM systems, RAG pipelines, security hardening, and prompt engineering for production agents.

triggers:multi-agentReActRAGprompt versioningfile lockingLLM securityplan-executeagent orchestration

GitHub SKILL.md

What it does

A practical engineering reference for building LLM-powered agents and multi-agent systems. Covers single-agent patterns (ReAct, plan-execute, reflection), multi-agent coordination (team creation, file locking, role decomposition), LLM security (prompt-injection defenses, jailbreak mitigation, output filtering), and RAG architecture (chunking, hybrid retrieval, reranking). Includes operational guidance: scale decisions, lifecycle rules, and integration points to provisioning and CI.

When to use it

Use this when designing or hardening production agent systems: choosing single vs multi-agent architectures, implementing RAG pipelines, establishing prompt/version control, or defining security/evaluation guardrails for deployed LLM apps. Helpful for architects, SREs, and lead engineers building agent orchestration.

What's included

Scripts: none bundled in this skill (has_references=true)
References: extensive reference docs for agent-dev, multi-agent coordination, protocol design, LLM security, and RAG systems (referenced under references/).
Instructions: scale-decision heuristics, multi-agent enablement criteria, file-locking and downgrade paths, and explicit rules for prompt versioning and I/O validation.

Compatible agents

Best suited for agent frameworks and tools that support composition and tool-calling (Claude Code, Codex-style agents, Gemini CLI, and OpenClaw orchestrations).

Audit Summary

Reference guide for building single- and multi-agent LLM systems covering ReAct, multi-agent coordination, LLM security, RAG, and prompt evaluation. Purely informational with no executable scripts. SKILL.md is well-structured with routing table and decision flowcharts, but content is primarily in Chinese which limits accessibility. No actionable automation — reads like a wiki rather than an invocable skill.

Watch Out

Content is primarily in Chinese — may limit usefulness for English-speaking users
No executable components — purely a reference document

Notes

Clean reference skill with no security concerns. Well-organized architecture with proper frontmatter and progressive disclosure. Main limitation is language (Chinese) and lack of actionable scripts — it's a knowledge reference, not an automation tool.

Information

Repository: code-abyss
Stars: 238

Trust Score

Overall78

Security100

Code Quality45

Architecture72

Usefulness55

More from code-abyss

Documentation Generator

Generates README.md and DESIGN.md scaffolds from a module's code and structure to accelerate project documentation.

Module Completeness Verifier

Scans directory structures to detect missing documentation and ensures synchronization between code and design documents.

Related Skills

ds-fix — data-science mid-analysis fixer

Orchestrates diagnosis and targeted fixes mid-analysis: diagnose root cause, apply fixes with output-first verification, and update project learnings.

Claw Bench

Benchmarking skill that guides an agent through a structured suite of capability tests and reporting steps for leaderboard submission.

Flutter & Firebase Integration Stack

Advanced integration patterns for Flutter, Firebase (Auth, Firestore, Functions), and Riverpod state management.

Spring Boot Migration

Guided, phased migration for existing Spring Boot apps to Boot 4 / Java 25 (including Modulith 2 and Testcontainers 2), with scanners and reference guides.

Java 25 & Spring Boot 4 Code Reviewer

Run focused, evidence-based code reviews for Java 25 and Spring Boot 4 projects — migration risks, architecture boundaries, null-safety (JSpecify), security, an

ArcKit Architecture Conformance

Run a systematic architecture conformance assessment that compares ADRs and principles against HLD/DLD artifacts to surface drift, violations, and technical deb

Neo4j Graph Database

Run Cypher queries and manage nodes/relationships against a local Neo4j instance exposed in the Docker network.

Brave Search CLI (bx)

High-performance web search and RAG grounding tool optimized for AI agents with token-budgeted content and custom ranking.

Back to Skills