🐇 Research Rabbit — Seed Expansion for PUMA Corpus
Tool: Research Rabbit (https://app.researchrabbit.ai) Phase: Phase 1 — Research Step: 01 — Literature Exploration Methodology: Citation Snowballing + SLR
Seed Papers (Primary Seeds for PUMA)
Use these verified papers as seeds in Research Rabbit:
- Assalaarachchi (2026) — Agentic SPM (arXiv:2601.16392)
- Cinkusz et al. (2025) — Cognitive Agents Agile PM (arXiv:2508.16678)
- Yao et al. (2022) — ReAct (arXiv:2210.03629)
- Jimenez et al. (2023) — SWE-bench (arXiv:2310.06770)
- MASAI (2024) — Arora et al. (arXiv:2406.11638)
Prompt A — Focused Seed Expansion
Build a focused research graph around AI agents for software project management, with emphasis on: issue triage, effort estimation, risk prediction, PMO automation, and evaluation benchmarks. Prioritize 2023–2026 papers, empirical studies, and reproducible frameworks. Use "Find Similar Papers" and "Who Cites This?" for all seed papers.
Prompt B — Gap Discovery via Graph
Identify underexplored clusters in the AI agents + project management literature graph. Focus on: governance mechanisms, human-in-the-loop validation, observability tools, sustainability (carbon footprint), and reproducibility protocols. Return papers connecting weakly represented subtopics to benchmark design.
Prompt C — Core Reading Set Prioritization
From the expanded collection, identify the 20 most central papers for a PUMA literature review. Separate into: (1) benchmark papers, (2) multi-agent orchestration papers, (3) project management applications, (4) evaluation methodology papers. Provide brief justification for each selection.
Prompt D — MOC-Oriented Clustering
Cluster the graph into thematic groups suitable for a Map of Content: literature exploration, scientific mapping, extraction methods, synthesis frameworks, hypothesis design, experiment design, and critical review. Export as BibTeX for Zotero import.
Workflow Integration
- Add seed papers → generate “Similar works” and “Cited by”
- Export graph as BibTeX/CSV → import to Zotero collection
PUMA-Literature-2026 - Tag results:
#multi-agent,#triage,#estimation,#benchmark,#governance - Forward to PRISMA screening: PRISMA-Log
PUMA Relevance
Research Rabbit revealed 12 papers that keyword search missed in the initial PUMA SLR, including several Cinkusz 2025 co-citations and Assalaarachchi 2026 predecessors. Critical for PUMA’s bibliometric completeness.