Uma Abordagem Orientada a Dados de Proveniência para Rastreabilidade de Workflows Agênticos
Resumo
Com o crescente uso de agentes de IA na especificação e automação de workflows, LLMs passaram a ser responsáveis por decidir sobre tarefas e dependências de dados. Entretanto, por serem suscetíveis a alucinações, esses modelos podem comprometer a validade dos resultados, tornando essenciais o rastreamento e a análise de suas decisões. A captura desses dados é desafiadora, pois o processo decisório frequentemente não é explícito. Neste artigo, propomos uma abordagem de captura de proveniência para workflows agênticos, a fim de permitir a análise das decisões dos agentes e do impacto dessas decisões no workflow gerado. Essa captura é baseada na instrumentação de frameworks de agentes e na representação conforme o padrão W3C PROV, oferecendo maior granularidade do que os trabalhos existentes.
Palavras-chave:
Proveniência Agêntica, Workflows agênticos, IA Responsável
Referências
Agno (2026). The runtime for agentic software. [link]. Acessado em Março de 2026.
Arize AI (2024). Openinference: Open standard for llm observability. [link]. Accessed: 2026-03-25.
Babuji, Y., Woodard, A., et al. (2019). Scalable parallel programming in python with parsl. In Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (Learning), PEARC ’19, pages 22:1–22:8. ACM.
BeeAI (2026). Agentic workflow framework. [link]. Acessado em Março de 2026.
Chang, C.-C. and Echizen, I. (2025). The chronicles of foundation ai for forensics of multi-agent provenance. arXiv preprint arXiv:2504.12612.
Cloud Native Computing Foundation (2024). Opentelemetry: Observability framework for cloud-native software. [link]. Accessed: 2026-03-25.
CrewAI (2026). The leading multi-agent platform. [link]. Acessado em Março de 2026.
Davis, D. B., Featherston, J., Fukuda, M., and Asuncion, H. U. (2017). Data provenance for multi-agent models. In 2017 IEEE 13th International Conference on e-Science (e-Science), pages 39–48.
de Oliveira, D. C. M., Liu, J., and Pacitti, E. (2019). Data-Intensive Workflow Management: For Clouds and Data-Intensive and Scalable Computing Environments. Synthesis Lectures on Data Management. Morgan & Claypool Publishers.
Deelman, E., Ferreira da Silva, R., Vahi, K., Rynge, M., Mayani, R., Tanaka, R., Whitcup, W., and Livny, M. (2021). The pegasus workflow management system: Translational computer science in practice. Journal of Computational Science, 52:101200. Funding Acknowledgments: NSF 1664162.
Freire, J., Koop, D., Santos, E., and Silva, C. T. (2008). Provenance for computational tasks: A survey. Comput. Sci. Eng., 10(3):11–21.
Friedman, S., Rye, J., LaVergne, D., Thomsen, D., Allen, M., and Tunis, K. (2020). Provenance-based interpretation of multi-agent information analysis.
Gómez-Carmona, O., Casado-Mansilla, D., de Ipiña, D. L., and García-Zubia, J. (2024). Human-in-the-loop machine learning: Reconceptualizing the role of the user in interactive approaches. Internet of Things, 25:101048.
Huang, L., Yu, W., Ma, W., Zhong, W., Feng, Z., Wang, H., Chen, Q., Peng, W., Feng, X., Qin, B., and Liu, T. (2025). A survey on hallucination in large language models: Principles, taxonomy, challenges, and open questions. ACM Trans. Inf. Syst., 43(2).
Missier, P., Belhajjame, K., and Cheney, J. (2013). The w3c prov family of specifications for modelling provenance metadata. In Proceedings of the 16th international conference on extending database technology, pages 773–776.
Moraes, J. V., Ferrari, C., Rosseti, I., and de Oliveira, D. (2025). Exploring evolutionary patterns: A jupyter notebook for discovering frequent subtrees in phylogenetic tree databases. Journal of Information and Data Management, 16(1):232–239.
Pimentel, J. a. F., Murta, L., Braganholo, V., and Freire, J. (2017). noworkflow: a tool for collecting, analyzing, and managing provenance from python scripts. Proc. VLDB Endow., 10(12):1841–1844.
Pinhanez, C., Cavalin, P., Sanctos, C., Grave, M., and Primerano, Y. (2025). The non-determinism of small llms: Evidence of low answer consistency in repetition trials of standard multiple-choice benchmarks.
Plötzky, F., Britz, K., and Balke, W.-T. (2025). A conceptual model for attributions in event-centric knowledge graphs. Data Knowledge Engineering, 159:102449.
Schick, T., Dwivedi-Yu, J., Dessí, R., Raileanu, R., Lomeli, M., Hambro, E., Zettlemoyer, L., Cancedda, N., and Scialom, T. (2023). Toolformer: language models can teach themselves to use tools. In Proceedings of the 37th International Conference on Neural Information Processing Systems, NIPS ’23, Red Hook, NY, USA. Curran Associates Inc.
Silva, V., de Oliveira, D., Valduriez, P., and Mattoso, M. (2018). Dfanalyzer: runtime dataflow analysis of scientific applications using provenance. Proc. VLDB Endow., 11(12):2082–2085.
Sirqueira, T. F. M., Viana, M. L., Cunha, F. J. P. D., Nunes, I., and Lucena, C. J. P. D. (2018). Data provenance in multi-agent systems: relevance, benefits and research opportunities. International Journal of Metadata, Semantics and Ontologies, 13(1):9–19.
Souza, R., Gueroudji, A., DeWitt, S., Rosendo, D., Ghosal, T., Ross, R., Balaprakash, P., and Da Silva, R. F. (2025). Prov-agent: Unified provenance for tracking ai agent interactions in agentic workflows. In 2025 IEEE International Conference on eScience (eScience), pages 467–473. IEEE.
Souza, R., Skluzacek, T. J., Wilkinson, S. R., Ziatdinov, M., and da Silva, R. F. (2023). Towards lightweight data integration using multi-workflow provenance and data observability. In IEEE International Conference on e-Science.
Wang, L., Ma, C., Feng, X., Zhang, Z., Yang, H., Zhang, J., Chen, Z., Tang, J., Chen, X., Lin, Y., Zhao, W. X., Wei, Z., and Wen, J. (2024). A survey on large language model based autonomous agents. Frontiers of Computer Science, 18(6):186345.
Zadra, N., Rizzoli, A., and Rota-Stabelli, O. (2024). Comprehensive phylogenomic analysis of zika virus: Insights into its origin, past evolutionary dynamics, and global spread. Virus Research, 350:199490.
Zhao, W. X. et al. (2023). A survey of large language models. arXiv:2303.18223, 1(2).
Arize AI (2024). Openinference: Open standard for llm observability. [link]. Accessed: 2026-03-25.
Babuji, Y., Woodard, A., et al. (2019). Scalable parallel programming in python with parsl. In Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (Learning), PEARC ’19, pages 22:1–22:8. ACM.
BeeAI (2026). Agentic workflow framework. [link]. Acessado em Março de 2026.
Chang, C.-C. and Echizen, I. (2025). The chronicles of foundation ai for forensics of multi-agent provenance. arXiv preprint arXiv:2504.12612.
Cloud Native Computing Foundation (2024). Opentelemetry: Observability framework for cloud-native software. [link]. Accessed: 2026-03-25.
CrewAI (2026). The leading multi-agent platform. [link]. Acessado em Março de 2026.
Davis, D. B., Featherston, J., Fukuda, M., and Asuncion, H. U. (2017). Data provenance for multi-agent models. In 2017 IEEE 13th International Conference on e-Science (e-Science), pages 39–48.
de Oliveira, D. C. M., Liu, J., and Pacitti, E. (2019). Data-Intensive Workflow Management: For Clouds and Data-Intensive and Scalable Computing Environments. Synthesis Lectures on Data Management. Morgan & Claypool Publishers.
Deelman, E., Ferreira da Silva, R., Vahi, K., Rynge, M., Mayani, R., Tanaka, R., Whitcup, W., and Livny, M. (2021). The pegasus workflow management system: Translational computer science in practice. Journal of Computational Science, 52:101200. Funding Acknowledgments: NSF 1664162.
Freire, J., Koop, D., Santos, E., and Silva, C. T. (2008). Provenance for computational tasks: A survey. Comput. Sci. Eng., 10(3):11–21.
Friedman, S., Rye, J., LaVergne, D., Thomsen, D., Allen, M., and Tunis, K. (2020). Provenance-based interpretation of multi-agent information analysis.
Gómez-Carmona, O., Casado-Mansilla, D., de Ipiña, D. L., and García-Zubia, J. (2024). Human-in-the-loop machine learning: Reconceptualizing the role of the user in interactive approaches. Internet of Things, 25:101048.
Huang, L., Yu, W., Ma, W., Zhong, W., Feng, Z., Wang, H., Chen, Q., Peng, W., Feng, X., Qin, B., and Liu, T. (2025). A survey on hallucination in large language models: Principles, taxonomy, challenges, and open questions. ACM Trans. Inf. Syst., 43(2).
Missier, P., Belhajjame, K., and Cheney, J. (2013). The w3c prov family of specifications for modelling provenance metadata. In Proceedings of the 16th international conference on extending database technology, pages 773–776.
Moraes, J. V., Ferrari, C., Rosseti, I., and de Oliveira, D. (2025). Exploring evolutionary patterns: A jupyter notebook for discovering frequent subtrees in phylogenetic tree databases. Journal of Information and Data Management, 16(1):232–239.
Pimentel, J. a. F., Murta, L., Braganholo, V., and Freire, J. (2017). noworkflow: a tool for collecting, analyzing, and managing provenance from python scripts. Proc. VLDB Endow., 10(12):1841–1844.
Pinhanez, C., Cavalin, P., Sanctos, C., Grave, M., and Primerano, Y. (2025). The non-determinism of small llms: Evidence of low answer consistency in repetition trials of standard multiple-choice benchmarks.
Plötzky, F., Britz, K., and Balke, W.-T. (2025). A conceptual model for attributions in event-centric knowledge graphs. Data Knowledge Engineering, 159:102449.
Schick, T., Dwivedi-Yu, J., Dessí, R., Raileanu, R., Lomeli, M., Hambro, E., Zettlemoyer, L., Cancedda, N., and Scialom, T. (2023). Toolformer: language models can teach themselves to use tools. In Proceedings of the 37th International Conference on Neural Information Processing Systems, NIPS ’23, Red Hook, NY, USA. Curran Associates Inc.
Silva, V., de Oliveira, D., Valduriez, P., and Mattoso, M. (2018). Dfanalyzer: runtime dataflow analysis of scientific applications using provenance. Proc. VLDB Endow., 11(12):2082–2085.
Sirqueira, T. F. M., Viana, M. L., Cunha, F. J. P. D., Nunes, I., and Lucena, C. J. P. D. (2018). Data provenance in multi-agent systems: relevance, benefits and research opportunities. International Journal of Metadata, Semantics and Ontologies, 13(1):9–19.
Souza, R., Gueroudji, A., DeWitt, S., Rosendo, D., Ghosal, T., Ross, R., Balaprakash, P., and Da Silva, R. F. (2025). Prov-agent: Unified provenance for tracking ai agent interactions in agentic workflows. In 2025 IEEE International Conference on eScience (eScience), pages 467–473. IEEE.
Souza, R., Skluzacek, T. J., Wilkinson, S. R., Ziatdinov, M., and da Silva, R. F. (2023). Towards lightweight data integration using multi-workflow provenance and data observability. In IEEE International Conference on e-Science.
Wang, L., Ma, C., Feng, X., Zhang, Z., Yang, H., Zhang, J., Chen, Z., Tang, J., Chen, X., Lin, Y., Zhao, W. X., Wei, Z., and Wen, J. (2024). A survey on large language model based autonomous agents. Frontiers of Computer Science, 18(6):186345.
Zadra, N., Rizzoli, A., and Rota-Stabelli, O. (2024). Comprehensive phylogenomic analysis of zika virus: Insights into its origin, past evolutionary dynamics, and global spread. Virus Research, 350:199490.
Zhao, W. X. et al. (2023). A survey of large language models. arXiv:2303.18223, 1(2).
Publicado
08/09/2026
Como Citar
MOURA DE SOUSA, Valesca; RAMA FIORINI, Sandro; DE OLIVEIRA, Daniel.
Uma Abordagem Orientada a Dados de Proveniência para Rastreabilidade de Workflows Agênticos. In: SIMPÓSIO BRASILEIRO DE BANCO DE DADOS (SBBD), 41. , 2026, São Carlos/SP.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 99-112.
ISSN 2763-8979.
DOI: https://doi.org/10.5753/sbbd.2026.249150.
