SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2606.10316 · 2026

TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning

Qi Liu, Yitong Zhou, Daoyu Wang, Mingyue Cheng, Shuo Yu, Qingyang Mao, Xiaoyu Tao, Qingchuan Li

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 5 functions out of this paper's own repositories and ran 3 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
fishsure/TabClaw — 3 of 5
FunctionStatusWhere it lives
SkillDistiller Ran fishsure/TabClaw/agent/multi_agent.py
pointer only (licence: NONE) · get_code("84cb97de42dcb4a7")
StepRecord Ran fishsure/TabClaw/agent/multi_agent.py
pointer only (licence: NONE) · get_code("d1b814386815fa45")
WorkflowRecord Ran fishsure/TabClaw/agent/multi_agent.py
pointer only (licence: NONE) · get_code("44df9f1adf77c905")
AgentExecutor Not yet run fishsure/TabClaw/agent/multi_agent.py
pointer only (licence: NONE) · get_code("99bd3f6cfa32d103")
MultiAgentExecutor Not yet run fishsure/TabClaw/agent/multi_agent.py
pointer only (licence: NONE) · get_code("677101b9dc0719e4")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Spreadsheets and tables are widely used representations for structured data analysis, but effective analysis still requires substantial manual effort and domain expertise. Recent large language model (LLM) agents can automate parts of this process, but they often provide limited transparency into intermediate decisions, rely on implicit assumptions, struggle with multi-table comparison, and repeat similar workflows without adapting to a user's preferences. This paper presents TabClaw, an open-source interactive AI agent for spreadsheet manipulation and table reasoning. Users upload CSV or Excel files and issue natural-language requests; TabClaw clarifies ambiguous intent, exposes an editable execution plan, streams a ReAct-style tool-using analysis loop, dispatches specialist agents for parallel multi-table reasoning, and synthesizes findings with explicit consensus and uncertainty markers. Beyond one-off analysis, TabClaw records completed workflows, extracts persistent user memory, distills reusable skills from repeated tool-use patterns, supports package-style skill import, and upgrades skills from negative feedback. Experiments on spreadsheet manipulation and table reasoning benchmarks show that TabClaw improves executable task completion and reasoning performance while preserving an

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2606.10316")
get_code_for_paper("2606.10316")
have("2606.10316")

Connect an agent — have() is free.