Details

Primerjava ogrodij za razvoj agentno usmerjenih sistemov
ID Jereb, Martin (Author), ID Lavbič, Dejan (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (1,23 MB)
MD5: 8B2C8A860F6C9A910577AD1D345AEACD

Abstract
V zadnjem obdobju se hitro razvijajo pristopi za gradnjo agentno usmerjenih sistemov, ki temeljijo na velikih jezikovnih modelih in omogočajo avtonomno delovanje agentov z uporabo orodij, spomina in orkestracije. V okviru diplomske naloge predstavimo tri pristope za razvoj tovrstnih sistemov na treh ravneh: n8n kot vizualno okolje brez pisanja kode, LangChain/LangGraph kot visoko prilagodljivo odprtokodno ogrodje ter Google ADK kot oblačno usmerjeno rešitev za produkcijska okolja. Poseben poudarek namenimo informacijski varnosti. Predstavimo največja varnostna tveganja po lestvici OWASP Top 10 za LLM-aplikacije iz leta 2025 ter preučimo, kateri zaščitni mehanizmi so na voljo razvijalcem. Rezultat naloge je analiza treh ogrodij in trije delujoči prototipi: detektor lažnih spletnih trgovin (n8n), orodje za pripravo življenjepisa in motivacijskega pisma (LangGraph) ter osebni Airbnb asistent (Google ADK). Ogrodja smo analizirali z vidika funkcionalnosti, kompleksnosti implementacije, varnostnih zmogljivosti in primernosti za produkcijska okolja. Analiza pokaže, da je n8n najprimernejši za hitro prototipiranje in vizualno gradnjo agentnih tokov brez poglobljenega programerskega znanja, da LangGraph zagotavlja največji nadzor nad stanjem in orkestracijo agentov pri zahtevnejših scenarijih ter da je Google ADK optimalna izbira za produkcijsko uvajanje v Googlovem oblačnem okolju. Ugotovili smo tudi, da nobeno od ogrodij ni varno samo po sebi, zato morajo razvijalci varnostne mehanizme implementirati premišljeno in sistematično ter se ne zanašati na vgrajene sisteme.

Language:Slovenian
Keywords:agentno usmerjeni sistemi, n8n, Google ADK, LangChain, LangGraph, OWASP Top 10 za LLM aplikacije 2025
Work type:Bachelor thesis/paper
Typology:2.11 - Undergraduate Thesis
Organization:FRI - Faculty of Computer and Information Science
Year:2026
PID:20.500.12556/RUL-184487 This link opens in a new window
COBISS.SI-ID:286276355 This link opens in a new window
Publication date in RUL:08.07.2026
Views:220
Downloads:96
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Comparison of Frameworks for the Development of Agentic Systems
Abstract:
Approaches to building agentic systems based on large language models have been developing rapidly, enabling autonomous agent behaviour through the use of tools, memory, and orchestration. This thesis presents three approaches to developing such systems across three abstraction levels: n8n as a visual low-code/no-code environment, LangChain/LangGraph as a fully customisable open-source framework, and Google ADK as a cloud-oriented solution for production environments. Special attention is given to information security. We present the most significant security risks according to the OWASP Top 10 for LLM Applications 2025 and examine which protective mechanisms are available to developers. The thesis presents an analysis of the three frameworks and three working prototypes: a fake online shop detector (n8n), a CV and cover letter generation tool (LangGraph), and a personal Airbnb assistant (Google ADK). The frameworks are analysed in terms of functionality, implementation complexity, security capabilities, and suitability for production environments. The analysis shows that n8n is best suited for rapid prototyping and visual construction of agentic workflows without deep programming knowledge, that LangGraph provides the greatest control over state and agent orchestration in complex scenarios, and that Google ADK is the optimal choice for production deployment within the Google Cloud environment. A key finding is also that none of the frameworks is secure by default and that developers must implement security mechanisms thoughtfully and systematically, without relying on built-in safeguards.

Keywords:agentic systems, n8n, Google ADK, LangChain, LangGraph, OWASP Top 10 for LLM Applications 2025

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back