zhang-ziao/SkillFlow-Task
SkillFlow-Task: SkillFlow Test Tasks: Harbor dataset on Hugging Face with 166 tasks. SkillFlow Test Tasks is the task repository used in the SkillFlow benchmark for evaluating lifelong skill discovery, skill revision, and cross-task procedural transfer in autonomous agents. The dataset contains…
Tasks
- Build an Excel compensation model workbook at:
- Build an Excel compensation model workbook at:
- Harbor Task: University Faculty Compensation Model
- Build a multi-year Excel compensation workbook for a university's full-time faculty.
- Harbor Task: Property Management Compensation Model
- Build a multi-year Excel compensation workbook for a property operations team.
- Harbor Task: Airline Crew Compensation Model
- Harbor Task: Construction Union Compensation Model
- You are assisting a cloud cost governance team in reconciling an archived PDF snapshot against the latest workbook.
- You are assisting a regional hospital planning office in reconciling an archived PDF snapshot against the latest workbook.
- You are assisting a retail supply planning team in reconciling an archived PDF snapshot against the latest workbook.
- You are assisting a university budgeting office in reconciling an archived PDF snapshot against the latest workbook.
- You are assisting infrastructure operations in reconciling an archived PDF snapshot against the latest workbook.
- You are assisting a hospital pharmacy team in reconciling an archived PDF snapshot against the latest workbook.
- You are assisting a port operations team in reconciling an archived PDF snapshot against the latest workbook.
- You are assisting an academic affairs office in reconciling an archived PDF snapshot against the latest workbook.
- You are a DevOps Performance Engineer supporting NovaCraft Engineering's CI/CD platform.
- You are a Telecom Field Service Planner supporting the Analyze gate for daily field work-order closure productivity.
- You are an Industrial Engineer supporting the Analyze tollgate of a DMAIC project for North Fulfillment Center LLS operations.
- You are an Industrial Engineer supporting the Brightland Processing Center.
- You are a Hospital Lab Improvement Specialist supporting the Analyze gate for clinical lab daily panel completion performance.
- You are a Healthcare Quality Analyst supporting Clearview Regional Hospital.
- You are a Supply Chain Analyst supporting Meridian Global Logistics.
- You are a SOC Operations Analyst supporting the Analyze gate for enterprise security alert triage throughput.
- You are a University IT Service Operations Analyst supporting the Analyze gate for student IT helpdesk ticket resolution throughput.
- You are supporting a cycle-count variance audit for inventory accuracy.
- You are supporting an outbound-operations audit for carton handoff accuracy.
- You are helping a retail pricing team verify promotional register performance.
- You are supporting a distribution-center receiving review for inbound grocery loads.
- You are supporting a returns-processing audit for disposition accuracy.
- You are supporting a service-desk SLA review for queue health.
- You are supporting an operations-compliance review for weekly consultant timesheets.
- You are supporting a yard-management review for trailer detention compliance.
- You are reviewing overnight clinic shift payment claims.
- You are reconciling participant release requests for a clinical trials payment team.
- You are auditing field service billing packets before payment.
- You are auditing fleet maintenance chargeback packets before payment.
- You are reconciling research stipend disbursement packets.
- You are reviewing speaker honorarium requests for an events team.
- You are helping a finance operations team review travel reimbursement claims.
- You are reviewing cold-chain charge requests from a warehouse receiving team.
- You are repairing /root/input.pptx , a lab methods slide with an embedded Excel dilution workbook.
- You are correcting /root/input.pptx , a process engineering slide with an embedded Excel catalyst-balance workbook.
- You are repairing /root/input.pptx , an environmental compliance slide with an embedded Excel workbook that contains both archived and…
- You are fixing /root/input.pptx , an investor board slide with an embedded Excel FX matrix.
- You are updating /root/input.pptx , a treasury review slide with an embedded Excel spot-rate grid.
- You are fixing /root/input.pptx , a channel rebate slide whose PPTX package contains two embedded Excel workbooks: one live and one…
- You are fixing /root/input.pptx , a procurement slide with an embedded Excel packaging-conversion workbook.
- You are updating /root/input.pptx , an operations slide with an embedded Excel slot-factor grid.
- You are refreshing the November 2025 refund reserve package for Atlas Cloud Commerce. Update the existing workbook template from the…
- You are closing November 2025 for Cedar Ridge Manufacturing. Build the accrual rollforward workbook from the prepared schedules and GL…
- You are a Senior Staff Accountant at Aurisic.
- You are covering the August 2025 close for LatticeWare. Build the deferred revenue workbook from the prepared schedules and GL checkpoints.
- You are completing the December 2025 contract liability package for Meridian Mobility Network. Update the existing workbook template from…
- You are preparing the September 2025 warranty reserve package for Northstar Appliances. Build the warranty reserve workbook from a…
- You are finishing the September 2025 project accounting package for Orchid Health Systems. Build the capitalized project cost rollforward…
- You are refreshing the October 2025 rebate reserve package for Peregrine Devices. Update the existing workbook template using the base…
- You are finalizing the October 2025 commission asset rollforward for Solstice Energy Software. Build the workbook by aligning the nested…
- Finish the clinic intake summary clinic intake template.hwpx with the values in patient intake.json and save the result to /root/clinic…
- Prepare the event announcement document event announcement template.hwpx using the values in event data.json , then save the result to…
- Complete the inventory status report inventory report template.hwpx using the values in inventory data.json , then save the result to…
- Complete the project proposal document project proposal template.hwpx using the values in project proposal.json , then save the result to…
- Revise the existing renewal playbook renewal playbook.hwpx using renewal update.json and followups.csv , then save the updated file to…
- Prepare the warehouse safety audit brief safety audit template.hwpx using both audit overview.json and corrective actions.json , then save…
- Update the HWPX supplier contact sheet supplier contact template.hwpx using the values in supplier contact.json , then save the finished…
- Fill in the training feedback sheet training feedback template.hwpx using the values in training feedback.json , then save the result to…
- You are supporting a community pulmonology clinic that is reviewing its refill-cycle policy for inhaled maintenance therapies.
- You are supporting a regional diagnostics network that is reviewing whether certain panels should stay on the 14-day replenishment cadence…
- You are supporting a retail pharmacy decision on auto-refill policy.
- You are supporting a home infusion program that is deciding whether to keep weekly deliveries or move some therapies to 14-day delivery…
- You are supporting a specialty pharmacy mailer program that is reconsidering how often it ships adherence refill packs.
- You are supporting an oncology supportive-care program that is reviewing whether to keep frequent cooler dispatches or move to larger…
- You are supporting a regional pathology operations team that is reviewing whether certain assays should stay on the small-kit cadence or…
- You are supporting a medication synchronization program that is reconsidering how often it ships adherence card packs.
- You are supporting a regional vaccination outreach team that is reviewing whether to keep smaller crate dispatches or move to larger…
- Media Revenue Co-Movement Analysis
- Cold-Chain and Last-Mile Cycle Correlation
- Capital Spending Co-Movement Analysis
- Housing Cycle Alignment Analysis
- Logistics and Warehousing Cycle Alignment
- Service Demand Synchronization Analysis
- Network Services Cycle Alignment
- Distribution and Packaging Cycle Comparison
- You are a sales and inventory planning analyst for a beverage distributor.
- You are a reliability planning analyst in manufacturing maintenance.
- You are the Fare Programs Analyst at MetroLink Transit Authority.
- You are a resilience planning analyst in datacenter backup-fuel planning.
- You are the Content Rights Operations Manager at Aurora Stream.
- You are a fulfillment planning analyst in perishable meal-kit distribution.
- You are the Cloud Capacity Operations Lead at Nimbus Compute.
- You are a workforce planning analyst in hospital operations.
- A cardiology workflow already created an output template file, and you need to replace it with harmonized data instead of creating a…
- An electrolyte monitoring export needs the same harmonization workflow as the other panels, but the mix of columns is different.
- You are cleaning a hepatology lab export that was merged from multiple hospitals.
- An ICU metabolic panel was split across two source files from different systems.
- A neonatal sepsis surveillance export uses mixed unit systems and inconsistent header names.
- An oncology follow-up export contains repeated draws for the same case, and you only need the most recent complete draw per case.
- A respiratory monitoring export is stored as JSON instead of a flat CSV.
- A thyroid clinic exported follow-up labs from different analyzers, and the units are inconsistent.
- A transplant monitoring panel is split across two lab exports with semantically related but not identical record keys.
- Task description
- Task description
- Task description
- Task description
- Task description
- Task description
- Task description
- Task description
- You are the Farm Operations Manager for a Manitoba agricultural harvest operations facility.
- You are the Data Center Operations Director for a Manitoba IT data center operations facility.
- You are a welding operations supervisor building a catch-up plan for MIG welding capacity.
- You are the Production Manager for a Manitoba automotive accessory plant.
- You are the Warehouse Operations Manager for a Manitoba warehouse fulfillment operations facility.
- SOC Alert Queue Catch-Up Plan
- Radiology Reading Backlog Stabilization
- Returns Center Throughput Ramp-Down
- Please clean up the floating archive captions in /root/Archive-Photo-Review.pptx .
- Please clean up the floating trail-stop captions in /root/Coastal-Trail-Guide.pptx .
- Please clean up the floating evidence captions in /root/Incident-Briefing.pptx .
- Please clean up the floating exhibit captions in /root/Museum-Exhibit-Board.pptx .
- Please clean up the floating location captions in /root/Property-Portfolio.pptx .
- Please clean up the floating site captions in /root/Storm-Damage-Survey.pptx .
- Please clean up the floating platform captions in /root/Transit-Platform-Review.pptx .
- Please clean up the floating species captions in /root/Wildlife-Field-Guide.pptx .
- You are a welding operations supervisor building a catch-up plan for MIG welding capacity.
- You are a welding operations supervisor building a catch-up plan for MIG welding capacity.
- You are a CNC operations supervisor building a catch-up plan for milling machine capacity.
- You are an electronics manufacturing supervisor building a catch-up plan for PCB assembly capacity.
- You are an HVAC installation supervisor building a staffing schedule for ductwork crew capacity.
- You are a glass manufacturing supervisor building a catch-up plan for furnace crew capacity.
- You are a textile manufacturing supervisor building a catch-up plan for dyeing process capacity.
- You are a shipbuilding operations manager building a catch-up plan for block assembly crew capacity.
- You are a chemical processing operations supervisor building a catch-up plan for reactor crew capacity.
- Create /root/answers.json from the filing data in /root/2025-q2 and /root/2025-q3 .
- Use the filing data in /root/2025-q2 and /root/2025-q3 together with /root/alerts input.json to create /root/answers.json .
- Update the existing JSON report at /root/brief template.json and save the completed result as /root/answers.json .
- Create /root/answers.json using the Q3 filing data in /root/2025-q3 .
- Prepare /root/answers.json by comparing one manager's stock holdings between /root/2025-q2 and /root/2025-q3 .
- Review the Q3 2025 filing data in /root/2025-q3 and prepare a single JSON file at /root/answers.json .
- Create /root/answers.json for an issuer ownership rollup using the 13F data in /root/2025-q2 and /root/2025-q3 .
- Create /root/answers.json from the filing data in /root/2025-q3 .
- Read the organizational hierarchy from /root/org hierarchy.pdf , expense transactions from /root/expense transactions.csv , and budget…
- Read the department directory from /root/departments.pdf and employee compensation data from /root/employee compensation.csv , then create…
- Read the event catalog from /root/event catalog.pdf , online registrations from /root/online registrations.xlsx , and walk-in…
- Read the product master list from /root/product info.pdf and the inventory records from two warehouse files: /root/warehouse a…
- Read the book catalog from /root/book catalog.pdf and the circulation records from /root/circulation records.xlsx , then create a new…
- Read the product catalog from /root/product catalog.pdf and the sales transaction data from /root/sales transactions.xlsx , then create a…
- Read the part specifications from /root/part specifications.pdf and inspection records from /root/inspection records.xlsx , then create a…
- Read the student roster from /root/student roster.pdf and course grades from /root/course grades.xlsx , then create a new report called…
- You are a bakery replenishment planner reviewing weekly rack coverage.
16 more tasks.