GITHUB AGENT SKILLS / 技能收录

给你的 AI 助手
装上专业技能

我们从 GitHub 持续收录、审核并分类优质的 Agent Skills,按职业与使用场景整理,复制一行命令即可安装到 Claude Code、Cursor 等智能体。

01 /

按职业浏览 Agent Skills

为不同岗位挑选的高频技能

全部职业
02 /

按使用场景浏览 Agent Skills

从你要完成的任务出发

全部场景
03 /

按创作者浏览 Agent Skills

按 GitHub Stars 排序的优秀创作者

全部创作者

全部 Skills 482556

Tashima-Tarsh
Tashima-Tarsh/Disha6.6

remote-sensing-scientist

Think and work like an expert Remote Sensing Scientist. Use when a task calls for Remote Sensing Scientist judgment. Reasons from sensor physics, atmospheric state, surface BRDF, and sampling geometry through Sen2Cor/LaSRC/6S atmospheric correction, sub-pixel coregistration, SAR radiometric terrain correction, and Olofsson area-adjusted accuracy while treating misregistration, NDVI saturation, BRDF anisotropy, mixed pixels, and spatial label leakage as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

reliability-engineer

Think and work like an expert Reliability Engineer. Use when a task calls for Reliability Engineer judgment. Reasons from failure mechanisms, time-to-failure distributions, censored field data, and stress-strength interference through FMECA, physics-of-failure models (Coffin-Manson, Arrhenius, Peck), Weibull and Crow-AMSAA growth analysis, and demonstration tests while treating mixture populations, wrong acceleration models, common-cause failures, and lab-pass-equals-field-proof as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

reinforcement-learning-researcher

Think and work like an expert Reinforcement Learning Researcher. Use when a task calls for Reinforcement Learning Researcher judgment. Reasons from MDP/POMDP structure, Bellman contraction and the deadly triad, and policy-gradient variance through Gymnasium 1.x/MuJoCo v5/ALE v5 protocols, CleanRL/SB3/JAX (MJX, MuJoCo Playground) stacks, rliable IQM with stratified bootstrap CIs, Minari offline datasets, and GRPO/RLVR post-training while treating truncation-as-termination bootstrap bugs, seed and hyperparameter selection bias, reward hacking, and offline extrapolation error as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

rehabilitation-scientist

Think and work like an expert Rehabilitation Scientist. Use when a task calls for Rehabilitation Scientist judgment. Reasons from ICF/disablement models, COSMIN MCID/MDC triangulation, TIDieR-Rehab/CONSORT 2025 trial design, gait lab and PROMIS outcomes, motor-learning mechanisms, and RE-AIM implementation science; treats natural recovery, therapist allegiance, and lab-vs-function confounds as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

regulatory-affairs-scientist

Think and work like an expert Regulatory Affairs Scientist. Use when a task calls for Regulatory Affairs Scientist judgment. Reasons from CTD/eCTD Modules 1–5 traceability, FDA Type B/EOP and EMA PRIME/scientific-advice strategy, ICH Q8–Q12 lifecycle CMC, and expedited pathways (BTD/RMAT/accelerated approval); treats RTF, clinical-hold CMC gaps, ignored meeting minutes, and eCTD validation failures as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

regenerative-medicine-scientist

Think and work like an expert Regenerative Medicine Scientist. Use when a task calls for Regenerative Medicine Scientist judgment. Reasons from potency assurance, 361 vs 351/ATMP pathways, USP <1043> ancillary tiers, G-Rex/closed CAR-T manufacture, MSC matrix potency, and ISO 10993/dECM scaffolds while treating comparability-without-bioassay and CFU-F-as-potency as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

reaction-engineering-specialist

Think and work like an expert Reaction Engineering Specialist. Use when a task calls for Reaction Engineering Specialist judgment. Reasons from rate laws, stoichiometry, residence-time distributions, and coupled heat-rate balances through LHHW and Michaelis-Menten kinetics, Thiele/effectiveness and Weisz-Prater diffusion criteria, tracer RTD models, and RC1/ARC calorimetry while treating thermal runaway, hot spots, catalyst deactivation, channeling, and k_L-a mass-transfer masking as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

radiologist

Think and work like an expert Radiologist. Use when a task calls for Radiologist judgment. Reasons from modality–question fit, contrast kinetics, and ACR Appropriateness Criteria (1–9); applies BI-RADS, LI-RADS, PI-RADS, and Lung-RADS with Fleischner/incidental-findings algorithms; integrates PACS/RIS/DICOM/IHE workflows, CTDIvol/DLP/DIR dose stewardship, and critical-results communication while treating perceptual misses, satisfaction of search, and CT/MRI/PET artifacts as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

radiochemist

Think and work like an expert Radiochemist. Use when a task calls for Radiochemist judgment. Reasons from radionuclide half-life, specific activity, radiochemical purity, and dosimetry through analytical/prep HPLC with radiodetector, iTLC, HPGe γ-spectroscopy, OLINDA/MIRD, and USP <823>/EANM release specs while treating defluorination, transchelation of ⁶⁸Ga/⁸⁹Zr, ⁹⁹ᵐTc colloid and ⁶⁸Ge breakthrough as first-class failure modes.

资料检索与研究
★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

radio-astronomer

Think and work like an expert Radio Astronomer. Use when a task calls for Radio Astronomer judgment. Reasons from the van Cittert–Zernike relation, the RIME measurement equation, the radiometer equation, and ν⁻²/λ² plasma propagation through CASA/WSClean/DDFacet imaging, Perley–Butler flux scaling, MT-MFS and w-stacking, RM synthesis, single-dish T_A*/T_mb calibration, and PRESTO/PSRCHIVE/PINT pulsar and FRB workflows while treating missing short spacings and negative bowls, self-calibration ghosts and flux suppression, RFI masquerading as dispersed bursts, polarization leakage and beam squint, and flaring flux calibrators as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

radiation-oncology-physicist

Think and work like an expert Radiation Oncology Physicist. Use when a task calls for Radiation Oncology Physicist judgment. Reasons from absorbed dose, fluence, beam geometry, and constraint-driven plan quality through TG-51/TRS-398 reference dosimetry, TPS engines (AAA, Acuros XB, Monte Carlo), DVH metrics, and gamma-based patient-specific QA while treating stale CT-to-density tables, couch-shift sign errors, MLC leaf-bank swaps, and small-field output mishandling as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

quaternary-scientist

Think and work like an expert Quaternary Scientist. Use when a task calls for Quaternary Scientist judgment. Reasons from dated landform-sediment-proxy associations, multi-method chronology, and ice-age cyclicity (MIS, orbital forcing) through radiocarbon/OSL/cosmogenic dating, Bayesian age models (OxCal, Bacon, IntCal20), tephrochronology, and GIA models while treating uncalibrated 14C years, incomplete OSL bleaching, cosmogenic inheritance, and no-analog pollen assemblages as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

quantum-physicist

Think and work like an expert Quantum Physicist. Use when a task calls for Quantum Physicist judgment. Reasons from Hilbert-space density operators, commutation relations, and Lindblad open-system dynamics through randomized benchmarking, gate-set and process tomography, Bell-CHSH tests, and Stim/PyMatching surface-code decoding while treating crosstalk, leakage, calibration drift, and measurement backaction as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

quantum-optics-scientist

Think and work like an expert Quantum Optics Scientist. Use when a task calls for Quantum Optics Scientist judgment. Reasons from field quadratures, atom-photon coupling (g, κ, γ), and heralding efficiency budgets through g⁽²⁾ Hanbury Brown-Twiss measurement, balanced homodyne tomography, HOM interference, and SNSPD/APD detector calibration while treating afterpulsing-faked antibunching, LO phase drift erasing squeezing, accidentals and dark counts, and unaddressed Bell-test loopholes as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

quantum-information-scientist

Think and work like an expert Quantum Information Scientist. Use when a task calls for Quantum Information Scientist judgment. Reasons from operational entropies (smooth min-entropy, coherent information), Stinespring/Choi channel pictures, and the achievability-versus-converse split through dual-certified SDP/NPA bounds, EAT/GEAT finite-key security proofs, Stim/PyMatching/BP+OSD circuit-level QEC simulation, classical shadows and GST, and ETSI/ISO QKD evaluation standards while treating asymptotic key rates passed off as performance, uncertified superadditivity numerics, code distance mistaken for circuit distance, MLE-tomography bias, and CHSH violations mistaken for security proofs as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

quantum-computing-scientist

Think and work like an expert Quantum Computing Scientist. Use when a task calls for Quantum Computing Scientist judgment. Reasons from qubits as noisy open systems through T1/T2, gate fidelity, RB/GST/XEB, and quantum volume to surface-code QEC; compiles with Qiskit/Cirq, applies ZNE/PEC/readout mitigation, and treats crosstalk, transpilation depth, and calibration drift as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

quantum-chemist

Think and work like an expert Quantum Chemist. Use when a task calls for Quantum Chemist judgment. Reasons from the Schrödinger equation through HF, MP2/CCSD(T)/CBS, and multireference (CASSCF/CASPT2); uses ORCA/Psi4/Gaussian with GMTKN55/WTMAD-4 validation, T1/D1 diagnostics, Helgaker CBS extrapolation, and BSSE/spin-contamination checks while treating SCF near-degeneracy, intruder states, and global-vs-local multireference masking as first-class failure modes.

★ 0
Tashima-Tarsh
Tashima-Tarsh/Disha6.6

quantitative-biologist

Think and work like an expert Quantitative Biologist. Use when a task calls for Quantitative Biologist judgment. Reasons from SBML/PEtab ODE models, structural and profile-likelihood identifiability, Bayesian inference (Stan/PyMC/AMICI), and live-cell pipelines (Cellpose/TrackMate/PhotoFiTT, REMBI); treats sloppiness, phototoxicity, and segmentation-tracking artifacts as first-class failure modes.

★ 0