oh-my-LibRPA

ABACUS + LibRPA GW workflow guidance and static input checks. Use when planning, preparing, or validating GW runs, including SCF/DF/NSCF chaining, librpa.in consistency, and safe run-directory setup.

Chat-first orchestrator for ABACUS + LibRPA workflows. Use when users ask in natural language to prepare, run, audit, debug, or regression-test GW/RPA tasks, especially when the agent must classify files, choose local vs server execution, route by system type, or verify new feature work without exposing raw CLI complexity.

Obtain, select, validate, or document ABFS_ORBITAL inputs for ABACUS + LibRPA workflows. Use when the user asks how to get .abfs files, how to write ABFS_ORBITAL entries in STRU, how to build or source auxiliary bases, or when the user is teaching a new ABFS_ORBITAL recipe that should be saved for reuse.

Stack-layer workflow for ABACUS -> LibRPA cases. Use when users provide ABACUS-style inputs such as INPUT/INPUT_scf/INPUT_nscf/KPT/STRU, or ask to run, validate, or debug LibRPA starting from ABACUS outputs. Route inside this layer to the existing GW, RPA, or Debug skills instead of mixing ABACUS conventions with FHI-aims workflows.

Use when users ask for periodic-solid FHI-aims + LibRPA g0w0_band workflows, especially when they need separate aims and LibRPA jobs, a control.in skeleton from geometry.in, or alignment to an ABACUS comparison case.

Use when the user asks to install, configure, compile, rebuild, or debug a FHI-aims build, especially when CMake reports missing ELSI content, MPI wrapper mismatches, or stale compiler and MKL cache paths.

Stack-layer workflow for FHI-aims -> LibRPA QSGW/G0W0 cases. Use when users ask to mirror existing FHI-aims + LibRPA cases, prepare staged k-point or basis campaigns, run aims before LibRPA, submit case-local Slurm scripts, or debug QSGW band workflows such as qsgw_band and qsgw_band0. Keep this layer separate from ABACUS INPUT/KPT/STRU workflows.

ABACUS + LibRPA RPA workflow guidance with conservative defaults and static checks. Use when preparing or troubleshooting RPA calculations.

Diagnose ABACUS + LibRPA RPA/GW failures from logs and inputs. Use when runs fail, outputs are inconsistent, or there are parser/read errors such as stod issues.