cs.AIOct 7, 2026
SaveRunningTab: Direct Workspace Interaction with Environment-Side Tabs
Organizations: KAIST
Abstract
Much knowledge work produces new deliverables from files a workspace already holds, and LLM agents are beginning to take such work over. Through direct corpus interaction, an agent can search and read any of those files from a terminal with no indexing, and producing a deliverable from many of them in this way is what we call direct workspace interaction (DWI). Reaching the files, however, is only half the task: nothing keeps track of what the task asks for, what has been read, and what was listed but never opened, all of which slip through the context window without leaving a trace, so an agent may extract a figure and still deliver a report without it. To address this, we present RunningTab, a framework that equips direct workspace interaction with an environment-side tab: a per-task record of what the task still owes, kept by the environment alongside the agent. Specifically, the agent adds its requirements, while the environment records every file read as an excerpt with its provenance and every listed but unopened file as a candidate; the agent can then see each requirement beside its best-matching excerpts and top unopened candidates, resolve it against matching content or set it aside with a reason, and, should it try to finish with requirements still open, receive them in a finish check. We validate RunningTab on three benchmarks with three LLMs, where it consistently outperforms plain DWI and baselines that keep the record in the model, while its tab usually holds the values a deliverable needs once seen.
Figures & tables
Figure 1 : Direct workspace interaction (DWI) and RunningTab on an illustrative task: a brief that needs three facts scattered across a workspace. (Left) In DWI, the agent reaches every file through its terminal, but what it reads lives only in its context window: facts read early are displaced by later observations, and the brief is delivered without them. (Right) RunningTab keeps an environment-side tab beside the agent: the agent adds the requirements of the task, the environment records what is read and what is listed but not opened, and the facts remain available until the deliverable is written.
Figure 2 : DCI with GPT-5.4 nano on Workspace-Bench, as shares (%). (Top) Lost points of the rubric pass rate by failure type. (Bottom) Rubric values shown to the agent, by outcome.
Figure 3 : RunningTab over one task. The agent (top) works in turns, and the environment-side tab (bottom) fills under the turn that wrote each entry. (A) The agent states the requirements. (B) As the agent lists and reads files, the environment records each listed file as a candidate and each read as a read log entry with its anchor. (C) The review shows each open requirement beside its best read log match and the unopened files. (D) The agent resolves a requirement by citing the read log entry that holds its content. (E) At the stop, the environment returns one finish check with what is still open; here the tab had already captured the missing fact, so the agent closes it and delivers.
| Workspace-Bench | TheAgentCompany | OfficeQA Pro | ||||
| Method | Pass Rate | TCR@70 | Score | Success | Acc@0% | Acc@1% |
| GPT-5.4 nano | ||||||
| DCI | 41.63 ( 0.90) | 20.67 ( 1.70) | 32.09 ( 2.22) | 17.86 ( 5.05) | 53.53 ( 3.27) | 65.38 ( 2.83) |
| TODO | 41.67 ( 0.95) | 23.33 ( 2.62) | 36.74 ( 2.74) | 22.62 ( 3.37) | 54.81 ( 3.14) | 65.71 ( 3.27) |
| Self-Refine | 41.32 ( 2.15) | 22.33 ( 3.09) | 35.76 ( 2.97) | 22.62 ( 4.45) | 54.17 ( 2.40) | 64.74 ( 2.40) |
| Self-Tracking | 42.93 ( 0.44) | 23.00 ( 1.41) | 34.30 ( 1.52) | 20.24 ( 1.68) | 55.13 ( 2.97) | 66.99 ( 3.17) |
Table 1 : Main results on three benchmarks with three LLMs, averaged over three runs (standard deviation in parentheses). Best results within each LLM are bolded ; second best are underlined .
Figure 4 : RunningTab with GPT-5.4 nano on Workspace-Bench (%). (Left) Values held in the tab, among those the agent was shown but its deliverable lacks. (Right) Held values that an LLM recovers when they are masked, by the part of the tab given.
Appendix figures & tables7 assets
Supplementary material from the paper’s appendix.
Appendix
| Task: Integrate the core information from four business travel management policy documents, covering the full workflow of travel approval, travel requirements, reimbursement standards, reimbursement procedures, and violation handling. Ensure the guide is clear, easy to understand, and compliant with company policy, and generate employee_business_travel_and_reimbursement_guide.doc . | |
| (A) Requirements | Turn 1: add , before any file is opened r1: travel approval workflow r2: travel requirements r3: reimbursement standards r4: reimbursement procedures r5: violation handling r6: generate employee_business_travel_and_reimbursement_guide.doc |
| (B) Candidates | Turn 3: find lists 98 files, which the environment records as candidates; after turn 8, list returns Unopened candidates: 94 (55 above threshold, marked *), ranked by relevance to the task; opening a file removes it 1. …/employee_travel_reimbursement_policy_2024.pdf * 2. …/travel_expense_reimbursement_standards_and_attachment_requirements.pdf * … |
| (B) Read log | Turns 5 to 8: read the four policy files, which the environment records as L1 to L4; show returns L2 as L2: …/business_travel_management_policy_2.txt (turn 6, 2286 chars) command: read …/business_travel_management_policy_2.txt --- Article 7 Standards for Reimbursement of Business Travel Expenses Employees traveling on business shall practice economy in transportation, meals, and … … Category City Category Accommodation Meal Allowance Long-distance Transportation … Manager level and above Tier-1 Cities 220 50 Bus or train (…) 25 … |
| Status line | Turns 3 and 8: appended to each observation [tab] … | candidates: 98 unopened , 59 above threshold | read log: 0 entries | … [tab] … | candidates: 94 unopened , 55 above threshold | read log: 4 entries | … |
| (C) Review | Turn 10: list Requirements: 6 open, 0 resolved r1, r2 [open] … r3 [open] reimbursement standards from your reads: L2 …: Article 7 Standards for Reimbursement of Business Travel Expenses … never opened: …/travel_expense_standards_and_reimbursement_rules_2024.docx * | … r4 to r6 [open] … |
| (D) Resolution | Turns 15 to 20: resolve , after writing the guide at turn 12 r1 done (L1) r2 done (L3) r3 done (L2) r4 done (L4) r5 done (L4) r6 unavailable (authored) |
Table 5 : The environment-side tab of RunningTab over one attempt on a Workspace-Bench task, by the stages (A) to (E) of Figure 3 . Boxes show what the agent writes to the tab (orange) and what the environment records or returns (blue), shortened with …; key parts for requirement r3 are in bold.
| Task: Based on the four newly downloaded water and electricity policy files, prepare a quarterly water-saving and power-saving communication and inspection campaign, compile a complete company_water_and_electricity_saving_management_manual.doc , and use it for company-wide guidance and quarterly inspections. … | ||
| DCI | RunningTab (Ours) | |
| (A) Requirements | No record of the task. | Turn 6: add r1: Analyze water_and_electricity_policy_1.txt r2 to r4: (the same for policy files 2 to 4) r5: Prepare a quarterly … campaign plan r6: Compile complete …_manual.doc in model_output r7: Output only the python list … |
| (B) Reads | Turns 3 to 6: read the four files (Chapters 1 to 5, Articles 1 to 20). | Turns 7 to 10: read the same files, recorded as L1 to L4 . |
| Writing | Turn 10: writes the manual. | Turn 19: writes the manual. |
| (C) Review | Turn 23: list r1 [open] Analyze water_and_electricity_policy_1.txt from your reads: L1 …policy_1.txt: Company Water and Electricity Conservation Policy … r2 to r7 [open] … | |
| (D) Resolution | Turns 24 to 30: resolve r1 done (L1) r2 done (L2) r3 done (L3) r4 done (L4) r5 to r7 unavailable (authored) | |
Table 6 : Case study on a Workspace-Bench task, comparing DCI and RunningTab with the same LLM by the stages (A) to (E) of Figure 3 . Boxes are colored as in Table 5 and shortened with …; the key parts are in bold.
| Task: Based on the purchase order on the desktop, a table is generated by supplier, summarizing the amount and purchase type … | ||
| DCI | RunningTab (Ours) | |
| (B) Reads | Turn 13: read …/Purchase_Order_8.txt Purchase Order #1008 Supplier: Supplier 15 … Amount: ¥55000 Status: Pending Approval | Turn 7: read the same order; the environment records it as L12 . L12: …/Purchase_Order_8.txt … Supplier: Supplier 15 … Amount: ¥55000 … |
| Writing | Turn 19: writes the table. | Turn 18: writes the table. |
| (E) Finish check | Turn 21: stops and delivers. | Turn 21: stops. [tab] Finish check 1 of at most 1: you stopped. No requirement items were recorded. Add one per fact, figure, or deliverable element the task names … |
| (A) Requirements | Turns 24 to 33: add r2: Table row data: Supplier 2 | Stocked In | ¥16000 … (r3 to r8) r9: Table row data: Supplier 15 | Pending Approval | ¥55000. Source: …/Purchase_Order_8.txt. … | |
| (D) Resolution | Turn 39: resolve r2 done (L8) … r9 done (L12) … r11 done (L9) | |
Table 7 : Case study on a second Workspace-Bench task, in the format of Table 6 : a value that DCI reads and leaves out of its table, and the same value in the tab, where the requirement for its row is resolved against the read-log entry of its purchase order. The gray box shows the purchase order as DCI reads it, and the blue box its read-log entry.
Figure 6 : Additions of RunningTab to the system prompt ( Section 3.3 ). Text in gray is the system prompt of the harness, shown only to locate the additions, with […] marking omitted parts.
Figure 7 : Operations on the tab ( Section 3.2 ), each given to the LLM as a tool with a description, followed by its parameters (marked ) with the values allowed or a description of each parameter.
Figure 8 : Messages from the environment during the task: (a) the status line, appended to each observation ( Section 3.3 ); (b) the review, returned by list with kind=requirements; and (c) the replies of resolve ( Section 3.4 ). In the review, entries and files are ranked by BM25 against the requirement, and each entry shows its first line sharing a term with it.
Figure 9 : Finish check, sent once at the first stop ( Section 3.4 ). A listed file is above the relevance threshold, and marked *, when its path shares at least two distinct terms with the text of the task.