Artifacts for Codlis: A Multi-Agent Framework for Interactive Source Code Visualization via Knowledge Graphs
Fuente:
Zenodo
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Recurso digital |
| Veröffentlicht: |
Zenodo
2026
|
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866901688910086144 |
|---|---|
| author | Anonymous |
| author_facet | Anonymous |
| contents | <p>Replication package for the ASE 2026 paper "Codlis: Code Listening -- An Interactive Multi-Agent System for Repository-Level Code Comprehension via Graph-Based Visualization."</p> <p>This artifact contains:</p> <ul> <li>(1) The complete source code for Codlis, including: <ul> <li>The Next.js frontend and Django backend with a five-agent LangGraph pipeline (Scanner, Analyzer, Architect, Illustrator, Proofreader) operating over a Neo4j knowledge graph.</li> <li>Docker Compose configuration for one-command deployment. </li> </ul> </li> <li>(2) Anonymized survey data and analysis scripts from a within-subjects crossover study with 10 software developers comparing Codlis against DeepWiki on code comprehension tasks across two open-source Python repositories (FastAPI and Flask)</li> <li>(3) Task designs and scoring rubrics used in the study</li> <li>(4) A video demonstration of the tool</li> </ul> <p>All participant data has been anonymized.</p> <p>See README.md for reproduction instructions.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_19245711 |
| institution | Zenodo |
| language | |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Artifacts for Codlis: A Multi-Agent Framework for Interactive Source Code Visualization via Knowledge Graphs Anonymous <p>Replication package for the ASE 2026 paper "Codlis: Code Listening -- An Interactive Multi-Agent System for Repository-Level Code Comprehension via Graph-Based Visualization."</p> <p>This artifact contains:</p> <ul> <li>(1) The complete source code for Codlis, including: <ul> <li>The Next.js frontend and Django backend with a five-agent LangGraph pipeline (Scanner, Analyzer, Architect, Illustrator, Proofreader) operating over a Neo4j knowledge graph.</li> <li>Docker Compose configuration for one-command deployment. </li> </ul> </li> <li>(2) Anonymized survey data and analysis scripts from a within-subjects crossover study with 10 software developers comparing Codlis against DeepWiki on code comprehension tasks across two open-source Python repositories (FastAPI and Flask)</li> <li>(3) Task designs and scoring rubrics used in the study</li> <li>(4) A video demonstration of the tool</li> </ul> <p>All participant data has been anonymized.</p> <p>See README.md for reproduction instructions.</p> |
| title | Artifacts for Codlis: A Multi-Agent Framework for Interactive Source Code Visualization via Knowledge Graphs |
| url | https://doi.org/10.5281/zenodo.19245711 |