718 skills · sorted by relevance
Integrate Flash Attention into transformer models to reduce processing time and GPU memory use for long sequences. Covers PyTorch integration, advanced attention options, and benchmarks that compare speed, memory use, and output accuracy.
Adapt PyTorch training scripts for multiple GPUs or machines with Accelerate. This skill guides code changes, configuration, and launch commands, including mixed precision, which combines numerical formats, and DeepSpeed or FSDP integration for distributed training.
Turn meeting notes, voice memos, photos, drawings, or requests for proposals into priced proposals and estimates. Use existing templates and historical rates to produce DOCX and PDF files, with approval-based signature routing and deposit invoicing where supported.
Read, create, and review PDFs with checks of both content and appearance. Uses Python tools for document generation and text extraction, then inspects rendered page images for spacing, alignment, and readability.
Improve GSAP animation performance by reviewing animated properties, browser layout work, and simultaneous updates. Use transforms, grouped updates, reusable animations, and targeted cleanup to reduce stuttering in lists, mouse followers, and scroll effects.
Triage incorrect MNN outputs and regressions by matching symptoms to focused debugging guides. Use backend comparisons, minimal reproductions, and controlled experiments to investigate memory issues, quantization, concurrency, numeric limits, caches, and inconsistent results across runs.
Adapt existing PyTorch training scripts to run across multiple GPUs or machines with Accelerate. This skill covers device setup, distributed training configuration, and mixed precision, which uses lower-precision numbers to reduce memory use.
Combine video clips first, then add background music or sound effects in a separate pass. The workflow checks source dimensions and duration, uses video copying when appropriate, and separates processing steps to avoid encoder compatibility problems.
Extract PDF text when the usual file reader returns errors or unusable output. Choose command-line tools or Python libraries for plain text, metadata, tables, and complex layouts, with checks for scanned or copy-protected documents.
Download PDFs, check that the files contain valid content, and try several text extraction tools in sequence. The workflow helps detect downloaded error pages, insufficient text, and extraction failures while recording diagnostic details.
Extract readable text from PDFs when normal file reading returns garbled or unusable content. The workflow switches between extraction tools, checks the results, and includes optical character recognition for scanned pages when needed.
Generate PDF checklists, reports, and assessments with sections, tables, and scoring criteria using Python. Choose a suitable library, save the document for download, and check the output, with shell execution as an alternative when the sandbox fails.
Extract PDF text using a sequence of tools when one method fails. Validate downloaded files, check whether the extracted text is substantial, record the successful method, and investigate alternative sources if extraction remains unsuccessful.
Build Solana apps with guidance for wallet connections, on-chain programs, testing, and deployment. Also covers balance and transaction lookups, troubleshooting, and migrating older code, with review steps before signing or sending transactions.
Classical end-to-end empirical analysis workflow in the traditional Stata ecosystem — native Stata + reghdfe + ivreg2 + csdid + did_imputation + eventstudyinteract + sdid + rdrobust + rddensity + synth + synth_runner + psmatch2 + teffects + ebalance + coefplot + esttab + asdoc + binscatter. **Defaults to economics empirical-paper style** (AER / QJE / AEJ) — every run produces a publication-ready output set with a multi-column regression table (M1→M6 progressive controls/FE) as the centerpiece, plus Table 1 (descriptives), mechanism / heterogeneity / robustness tables, and event-study + coefficient + trend figures. Covers the full 8-step Stata pipeline an applied economist runs on every paper — (1) data import & cleaning (use/import, destring, misstable, duplicates, merge assert), (2) variable construction (gen/egen/winsor2/xtile/xtset with L./F./D.), (3) descriptive statistics & Table 1 (tabstat/balancetable/asdoc), (4) classical diagnostic tests (sktest/swilk/hettest/imtest/xtserial/xttest3/vif/dfuller/kpss/hausman/estat overid), (5) baseline modeling (reg/xtreg/reghdfe/ivreg2/ivregress/csdid/did_imputation/eventstudyinteract/sdid/rdrobust/synth/psmatch2/teffects/heckman/qreg/ppmlhdfe), (6) robustness battery (bacondecomp/honestdid/rwolf/ritest/wildbootstrap/oster), (7) further analysis (subgroup/triple-diff/interactions/medsem/marginsplot/binscatter by group), (8) publication-ready tables & figures (esttab/outreg2/estout/coefplot/marginsplot/rdplot/twoway combined). **Also covers two parallel domain modes that share the same 8-step scaffolding** — **Mode A — Epidemiology / public health** (target-trial emulation, IPTW + g-formula + TMLE doubly-robust triplet via `teffects ipw` / `teffects ipwra` / `teffects aipw` / `eltmle`, Mendelian randomization via `mrrobust` (IVW / Egger / weighted median) and `mregger` / `mrpresso`, KM / Cox / AFT / RMST survival via `sts` / `stcox` / `streg` / `strmst2`, E-value sensitivity via `evalue` (Linden-Mathur), principal stratification — STROBE / TRIPOD reporting), and **Mode B — ML causal inference** (DML via `ddml` / `pdslasso`, S/T/X/R/DR meta-learners via `crforest` and `ddml interactive`, causal forest via `crforest` / `cforest`, BART/BCF via `bart` / `bartCause`-style externals, CATE distribution + policy tree via `crforest`, off-policy evaluation, conformal causal externals, fairness audit, DAG learning via `pcalg` / external Python callouts). Use when the user asks for a complete Stata empirical analysis, wants a reproducible .do-file pipeline, needs a Stata counterpart to the Python StatsPAI / Full-empirical-analysis-skill, or names a specific Stata step in isolation ("run reghdfe with two-way clustering", "csdid event study", "winsor2 at 1%", "esttab to LaTeX", "coefplot with CI", "ivreg2 weak-IV test", "synth_runner placebos", "teffects psmatch balance check"). Mode A triggers on "target trial emulation Stata", "teffects ipw aipw", "eltmle", "mrrobust", "mregger weighted median", "stcox AFT survival", "strmst2", "evalue Stata", "STROBE Stata", "公共健康 Stata", "流行病学 Stata". Mode B triggers on "ddml Stata", "pdslasso", "crforest causal forest Stata", "policy tree Stata", "因果机器学习 Stata".
The English summary is being prepared. Showing the original description.
Calculates molecular fingerprints (ECFP/Morgan, FCFP, MACCS, RDKit, AtomPair, TopologicalTorsion, Avalon, MAP4, MHFP6) and physicochemical descriptors (Lipinski, QED, TPSA, Crippen LogP, 3D shape) with explicit choice tables, bit vs count semantics, and partial-charge model selection. Use when featurizing molecules for similarity, QSAR, virtual screening, or ML, or selecting the correct fingerprint for a chemotype-aware task.
The English summary is being prepared. Showing the original description.
Bevy genre router for FPS shooter. Keywords: Bevy, FPS shooter, genre, ECS, router.
The English summary is being prepared. Showing the original description.
Thin Bevy genre router for FPS shooter. Keywords: Bevy, FPS shooter, genre, ECS, router.
The English summary is being prepared. Showing the original description.
提供在WinForms中通过ElementHost嵌入DiffPlex的WPF DiffViewer控件，实现两个字符串的对比及差异高亮显示的代码方案。
The English summary is being prepared. Showing the original description.
Think and work like an expert FPGA Engineer. Use when a task calls for FPGA Engineer judgment. Reasons from metastability budgets, setup/hold margins, and tool-reported WNS/TNS through Vivado/Quartus timing and CDC reports, XDC/SDC constraints, Spyglass/Verilator lint, and ILA/IBERT lab bring-up while treating unsafe clock-domain crossings, reset domain crossings, sim-versus-silicon X mismatches, and unconstrained timing paths as first-class failure modes.
The English summary is being prepared. Showing the original description.
Think and work like an expert Fire Protection Engineer. Use when a task calls for Fire Protection Engineer judgment. Reasons from NFPA 13 Hazen-Williams hydraulics (K-factor, remote area, hose stream) and NFPA 101 egress (occupant load, travel distance, capacity factors) through NFPA 92 smoke containment/management, ASET/RSET PBD, and FDS/CFAST/CONTAM/PyroSim modeling while treating breached compartmentation, C-factor/fitting errors, and supply-curve shortfall as first-class failure modes.
The English summary is being prepared. Showing the original description.
Use when the user wants to turn a static scene, character images, aerial map, drawn path, or route-control image into an immersive first-person FPV AI video prompt, especially Seedance/Kling/Runway/Veo style one-shot videos with numbered stop markers, red-line path control, world-map flythroughs, camera route planning, variable character counts, non-human POVs such as drones, pets, robot vacuums, character references, timed interactions, dialogue, spatial audio, and negative constraints.
The English summary is being prepared. Showing the original description.
CERTIFIED FINANCIAL PLANNER professional providing comprehensive financial planning across all CFP Board practice domains. Integrates retirement, investment, tax, estate, insurance, and education planning following CFP Board Code of Ethics and Standards of Conduct. Use for holistic financial plans, cash flow analysis, net worth statements, insurance needs, estate planning coordination, education funding (529 plans), tax strategies, retirement projections, or CFP Board Practice Standards compliance.
The English summary is being prepared. Showing the original description.
Calculates molecular fingerprints (ECFP/Morgan, FCFP, MACCS, RDKit, AtomPair, TopologicalTorsion, Avalon, MAP4, MHFP6) and physicochemical descriptors (Lipinski, QED, TPSA, Crippen LogP, 3D shape) with explicit choice tables, bit vs count semantics, and partial-charge model selection. Use when featurizing molecules for similarity, QSAR, virtual screening, or ML, or selecting the correct fingerprint for a chemotype-aware task.
The English summary is being prepared. Showing the original description.