964 件 ・ 関連度順
概要と使いどころ
Optimize DSPy programs with dspy.GEPA — a reflective/evolutionary optimizer to consider against task-specific baselines within an authorized evaluation budget. Use when the user says optimize, compile, GEPA, reflective optimization, or "make this program better" and a DSPy program + metric + trainset exist.
日本語の概要は準備中です。原文の説明を表示しています。
Build DSPy 3.2.x programs through spec, program, metric and baseline; extend to optimization and export when requested and justified by task budget. Orchestrates the other four DSPy skills (dspy-fundamentals, dspy-evaluation-harness, dspy-gepa-optimizer, dspy-rlm-module) in the correct order. Use for greenfield DSPy builds; prototypes may stop at a validated baseline.
日本語の概要は準備中です。原文の説明を表示しています。
Full Ads Funnel Architect. Builds a complete advertising funnel across TOFU, MOFU, BOFU, and Retargeting stages with platform-specific templates, audience targeting, ad formats, copy angles, landing page requirements, KPIs, and budget allocation for Meta, Google, and LinkedIn.
日本語の概要は準備中です。原文の説明を表示しています。
Full Ad Strategy Orchestrator. Launches 5 parallel subagents to build a complete advertising strategy from a single URL — audience personas, creative concepts, funnel architecture, competitive intelligence, and budget allocation. Produces a composite Ad Readiness Score (0-100) with a unified, client-ready strategy report.
日本語の概要は準備中です。原文の説明を表示しています。
AI creative director with recursive self-assessment. Generates concepts using world-class methodologies (SIT, TRIZ, Lateral Thinking, bisociation), scores against 6 weighted criteria with Cannes/D&AD/HumanKind calibration, and recursively refines until the 9+ threshold is reached. Accepts briefs in any format — text, voice transcript, PDF, or raw notes. Use when the user asks to generate creative concepts, brainstorm campaign ideas, develop a Big Idea or campaign platform, evaluate or critique existing creative work, find consumer insights, or shares a brief for ideation — including activations, PR-stunts, brand utility, experiential, and non-advertising ideas. Calibrates against a library of 569 legendary campaigns (P01-P18 pattern map) to detect saturation and ensure originality. Do not use for media planning, production budgeting, brand identity/logo design, copywriting final drafts, or market research data collection.
日本語の概要は準備中です。原文の説明を表示しています。
Define production-ready visual intent for Blender scenes and standalone assets, including focal hierarchy, depth layers, visual flow, scale and repetition rhythm, camera mobility, representation fidelity, and performance budgets. Use for original or reference-guided Blender work before construction choices are executed, especially environments, architecture, products, hero assets, or scenes whose modeling quality depends on coordinated spatial design rather than isolated objects.
日本語の概要は準備中です。原文の説明を表示しています。
Plan and optimize resource allocation. Trigger with "resource planning", "capacity", "utilization", "staffing plan", "who should work on what", "we're stretched thin", or when the user needs help allocating people, budget, or time across projects and teams.
日本語の概要は準備中です。原文の説明を表示しています。
Plan comprehensive marketing campaigns from strategy through execution. Develop campaign briefs, select optimal channels, create timelines, allocate budgets, define targeting parameters, design A/B testing strategies, and establish performance measurement frameworks.
日本語の概要は準備中です。原文の説明を表示しています。
Apple Human Interface Guidelines as structured design values — system colors (incl. increased-contrast variants), the full Dynamic Type ramp, Liquid Glass rules and budgets, layout and hit-target metrics, SF Symbols, and HIG-to-SwiftUI component mappings. Use when styling or reviewing any Apple-platform UI: picking a color, size, spacing, corner radius, or hit target; writing .glassEffect / material code; choosing an SF Symbol; or answering 'what does the HIG say about X'.
日本語の概要は準備中です。原文の説明を表示しています。
Compress selected context to a target token budget while preserving decisions, evidence, constraints, and unresolved questions. Use when relevant material is already selected but too long; use context-optimization when selection, deduplication, and ordering are also required.
日本語の概要は準備中です。原文の説明を表示しています。
Optimize machine learning model hyperparameters using grid search, random search, Bayesian optimization, and Hyperband to maximize model performance within a compute budget. Use when the user requests hyperparameter tuning or provides relevant inputs for this workflow.
日本語の概要は準備中です。原文の説明を表示しています。
The research-domain clarification skill — find underspecified items in a thesis spec.md (prose wrong_if, universe rows without rationale, missing budget/expiry/pins, ambiguous pillars), ask the human structured questions WITH options (max 5 per round), encode the answers back into spec.md's Clarifications section, and re-evaluate checklists/thesis-quality.md after every write (Q32 bidirectional maintenance).
日本語の概要は準備中です。原文の説明を表示しています。
Execute research tasks — checklist soft gate, phase dispatch with explicit thesis_dir, budget enforcement (halt + approval card on overrun), skill_pin recording (version_hash content-hashed per skill directory), [x] written on completion.
日本語の概要は準備中です。原文の説明を表示しています。
Scaffold and amend the L1 Investment Constitution — [ALL_CAPS] placeholder bootstrap, SemVer bump rules, Sync Impact Report, MINOR/MAJOR re-examination dispatch after the gate-5 budget confirm. Thesis creation stays refused until ratification (Q83).
日本語の概要は準備中です。原文の説明を表示しています。
position sizing, Kelly criterion, risk budgeting, volatility targeting, portfolio allocation, bet sizing, capital allocation, drawdown management, position management, risk parity sizing
日本語の概要は準備中です。原文の説明を表示しています。
Think and work like an expert Mechanical Design Engineer. Use when a task calls for Mechanical Design Engineer judgment. Reasons from function, datum reference frames, and tolerance budgets; releases inspectable drawings and MBD through ASME Y14.5 GD&T, WC/RSS/Monte Carlo stack-ups, SAE J1739 DFMEA (Action Priority), and Boothroyd–Dewhurst DFM/DFA—not stress plots alone.
日本語の概要は準備中です。原文の説明を表示しています。
Think and work like an expert Low-Temperature Physicist. Use when a task calls for Low-Temperature Physicist judgment. Reasons from kT budgets, He-3/He-4 dilution refrigeration, and BCS/GL superconductivity; measures Tc, QHE, and Landauer conductance with lock-in/SQUID workflows while treating wiring heat loads, Kapitza resistance, flux trapping, TLS dielectric loss, and sample-vs-MXC thermometer mismatch as first-class failure modes.
日本語の概要は準備中です。原文の説明を表示しています。
Think and work like an expert Metrology Scientist. Use when a task calls for Metrology Scientist judgment. Reason from SI traceability, GUM uncertainty budgets, and VIM distinctions between calibration and verification; propagates Type A/B components, CIPM MRA equivalence, and ILAC decision rules before any pass/fail claim.
日本語の概要は準備中です。原文の説明を表示しています。
Think and work like an expert Laser Physicist. Use when a task calls for Laser Physicist judgment. Reasons from population inversion and threshold gain (g ≥ l/L + T), ABCD resonator stability (g₁g₂ in 0–1) and c/(2L) mode spacing, GVD/TOD dispersion management and time-bandwidth ΔtΔν ≈ 0.44, and Kerr/SPM nonlinear phase through Kerr-lens/SESAM mode-locking, CPA stretcher–amplifier–compressor with B-integral budgeting, QPM in PPLN/PPLKTP for SHG/OPA, FROG/SPIDER and Dazzler pulse shaping, M² caustics per ISO 11146, and IEC 60825 classification while treating thermal-lens drift out of the stability zone, autocorrelation-width-mistaken-for-pulse-width, uncompensated TOD wings and regen double-pulsing, and LMA-fiber mode instability and SRS/SBS above kW as first-class failure modes.
日本語の概要は準備中です。原文の説明を表示しています。
Think and work like an expert Instrumentation Engineer. Use when a task calls for Instrumentation Engineer judgment. Reasons from the process variable, its transduction chain, 4-20 mA loop integrity, and traceable calibration through ISA-5.1 P&IDs, NAMUR NE43 fault bands, ISO 5167 orifice sizing, and GUM uncertainty budgets while treating plugged impulse lines, double-applied square-root, ground-loop noise, and unrevalidated SIS bypasses as first-class failure modes.
日本語の概要は準備中です。原文の説明を表示しています。
Think and work like an expert Glaciologist. Use when a task calls for Glaciologist judgment. Reasons from mass-budget closure (SMB, dynamic discharge, calving), Glen flow law, and subglacial effective pressure through WGMS/GlaMBIE stake networks, ICESat-2/CryoSat altimetry with firn and radar-penetration corrections, ITS_LIVE velocities, ApRES basal melt, RES/MCoRDS bed picks, RGI/BedMachine inventories, OGGM mountain-glacier projections, and PISM/ISSM ISMIP6/7 protocols while treating firn-compaction aliasing, DEM penetration bias, GRACE leakage/GIA, and tidal InSAR artifacts as first-class failure modes.
日本語の概要は準備中です。原文の説明を表示しています。
Think and work like an expert Electronics Engineer. Use when a task calls for Electronics Engineer judgment. Reasons from datasheet-corner device physics, signal-chain error budgets, and analog-digital return-path coupling through LTspice/IBIS-AMI simulation, ICT/boundary-scan coverage, golden-board signature comparison, and IPC/AEC-Q standards while treating ESD versus EOS overstress, MLCC DC-bias derating, reference and clock-jitter ENOB loss, and NFF field returns as first-class failure modes.
日本語の概要は準備中です。原文の説明を表示しています。
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.
日本語の概要は準備中です。原文の説明を表示しています。
Think and work like an expert Ethologist. Use when a task calls for Ethologist judgment. Reasons from Tinbergen's four questions and versioned species vs experimental ethograms; scores with BORIS (Cohen's κ per behavior), Altmann focal/scan budgets, and ARRIVE 2.0/study-plan lab reporting while treating observer expectation, field–lab arena mismatch, habituation, and pseudoreplication as first-class failure modes.
日本語の概要は準備中です。原文の説明を表示しています。