vibrantlabsai/pool-desk-v2
pool-desk-v2: Harbor dataset on Hugging Face with 24 tasks. An OpenEnv environment for the OpenEnv Arena: 24 terminal tasks built on real public data, covering all eight Arena domains. It is an experiment-designed subset of the same reviewed task pool as Pool Desk. Each episode is one task in a…
Tasks
- Tier the CVEs in a scanner export by CISA KEV (as of a past review date) ransomware use and membership, EPSS score and EPSS percentile…
- Count, per endpoint, the Foxit PDF vulnerabilities whose flattened NVD CPE applicability rows (version bounds, exact versions, platform)…
- Write a Python regex detection rule for failed SSH password attempts with named user and ip groups, scored on hidden real sshd lines.
- Compute the Texas ratio for 30 AZ/NV banks from Call Report items identified by caption, with RCFD/RCON selection and a 10% watch flag.
- Pick face-of-balance-sheet XBRL facts from SEC FSDS num.txt and report or derive total liabilities for 70 small-filer 10-Qs.
- Pick face-of-statement XBRL facts from SEC FSDS num.txt for 30 small-filer 10-Qs, compute current ratio and nine-month cash runway, and…
- Reconcile EPA CAMPD unit heat input with EIA-923 plant fuel consumption for one state's 2024 fleet: total vs electric fuel, nuclear…
- 2024 capacity factors for one state's thermal generators: EIA-923 monthly net generation joined to the EIA-860 inventory, with…
- Group FRA Form 54 rail accident reports into accidents through shared incident numbers and joint-report links, total the deduplicated…
- Estimate households and mean adjusted household income for three areas from ACS PUMS housing records, with successive-difference replicate…
- Crude and directly age-adjusted weighted smoking prevalence for three disguised states from BRFSS 2023 respondent records.
- Find curves filed under the wrong isogeny class in a re-modelled table of Cremona elliptic curves, by comparing a p from point counts.
- Compute MusicBrainz disc IDs and playing times from ripper LBA track tables (CD TOCs from MusicBrainz, jittered).
- Parse free-text museum dimension lines (inches with fractions, cm or mm), convert to cm and flag mistyped metric conversions.
- Reconcile a label's CD releases between MusicBrainz and Discogs exports by barcode and report track-count and runtime status per release.
- Compute per-storm accumulated cyclone energy for the 2024 global season from an IBTrACS-format extract (units line, synoptic-only fixes…
- Apply an ordered QC screen (duplicate timestamps, range checks that skip sentinels, gust below wind) to a year of NDBC buoy observations…
- Compute UniProt sequence coverage by PDB entries at or below 2.5 A by joining a SIFTS segment table to the wwPDB resolution index and…
- Build a May-to-June 2026 headcount bridge per sub-agency from OPM employment snapshots, accessions and separations, with grouped flow…
- Flag broken reporting lines (dangling, cycles, grade inversions, unrooted chains) with ordered rules in a UK department senior-post…
- Harmonise three UK departments' spending-over-£25k files (encodings, £ strings, schemas) and total spend for 14 suppliers matched by a…
- Resolve effective GITHUB TOKEN permissions for every job in 89 real workflows (job overrides workflow, defaults) and flag write-capable…
- Class 271 real PEP 508 requirement strings of 30 PyPI packages as pinned, bounded, floor or open (PEP 440 operator semantics) and count…
- Build a per-session ledger (client, reattaches, expiry vs client close, lifetime) from a real three-server ZooKeeper log where socket…