「eval」の検索結果

Plan, execute, document, and retest authorized security assessments of AI agents and multi-agent workflows using safe adversarial cases, synthetic identities, canaries, and evidence-based findings. Use when defining red-team rules of engagement, assessing prompt injection or excessive agency, testing tool and identity boundaries, evaluating memory or cross-agent attacks, scoring a campaign, or verifying remediation in an approved environment.

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。

Place trusted contextual information into prompts or agent state using explicit boundaries, provenance, and templates. Use when relevant context has already been selected and must be inserted safely; use context-retrieval to find it or context-optimization to choose and order it.

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。

Design, implement, and validate reproducible machine-learning pipelines spanning data preparation, training, evaluation, registry, and deployment gates. Use when the user requests an ML pipeline, needs to turn model scripts into an orchestrated workflow, or provides pipeline components that must be connected safely.

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。

Audit agent skills, plugins, prompts, manifests, scripts, dependencies, and bundled assets for provenance, prompt-injection, permission, execution, exfiltration, persistence, and update risk. Use when evaluating a third-party skill before installing, enabling, updating, publishing, or distributing it; reviewing an untrusted SKILL.md, agent configuration, MCP integration, archive, or repository; comparing a package with a known-good version; or investigating unexpected tool, network, credential, or filesystem behavior.

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。

Retrieve relevant information from a knowledge base using semantic, keyword, or hybrid search to ground a query. Use when the task starts with a corpus or index that must be searched; use context-ranking when candidate chunks already exist and only need ordering.

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 MLOps Engineer. Use when a task calls for MLOps Engineer judgment. Reasons from data contracts, feature parity, evaluation gates, and rollback-readiness through MLflow/W&B registries, Feast feature stores, KServe/Triton serving, Great Expectations/TFDV validation, and Evidently PSI/KS drift monitors while treating train-serve skew, data leakage, silent degradation, and schema/concept drift 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 Machine Learning Engineer. Use when a task calls for Machine Learning Engineer judgment. Reasons from decision-policy framing, Google's Rules of ML, point-in-time data, and prefill/decode inference physics through GBDT/PyTorch baselines, vLLM/SGLang/KServe serving with FP8/AWQ quantization, hybrid-retrieval RAG, RAGAS and human-validated LLM judges, OpenTelemetry GenAI tracing, and post-Omnibus EU AI Act obligations while treating train-serve skew, temporal leakage, prompt injection, LLM nondeterminism, and degenerate feedback loops 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 Computer Graphics Researcher. Use when a task calls for Computer Graphics Researcher judgment. Reasons from the rendering equation, Monte Carlo bias/variance, sampling theory, and BSDF energy conservation through pbrt-v4/Mitsuba 3 references, MIS/ReSTIR/path guiding, DXR 1.2 real-time stacks, 3DGS and NerfBaselines protocols, FLIP/ColorVideoVDP evaluation, OpenUSD/MaterialX/OpenPBR interchange, and ACES 2/OCIO color while treating unconverged references, unequal-time comparisons, denoiser and temporal-reuse bias, scene- vs display-referred color errors, and white-furnace energy failures 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 Conservation Biologist. Use when a task calls for Conservation Biologist judgment. Reasons from IUCN Red List A–E and Green Status recovery metrics, PVA/Ne, occupancy and distance sampling (unmarked, msocc, RMark), prioritizr/Marxan SCP, Conservation Evidence and ROSES synthesis, counterfactual impact evaluation, METT/SMART PAME, and eDNA false-positive models while treating pseudoreplication, GBIF effort bias, offset baselines, and Red List≠priority conflation 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 Computer Architecture Researcher. Use when a task calls for Computer Architecture Researcher judgment. Reasons from ISA semantics, AMAT/CPI, MESI coherence, and branch prediction through gem5/SPEC/MLPerf evaluation, Amdahl and roofline discipline, TPU/GPU dataflow accelerators, DVFS/EDP, and Spectre/Meltdown mitigation overhead at ISCA/MICRO/HPCA rigor.

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 Carbon Cycle Scientist. Use when a task calls for Carbon Cycle Scientist judgment. Reasons from carbon mass balance across reservoirs and timescales, sign conventions, and budget closure through the Global Carbon Budget protocol, ONEFlux/REddyProc eddy-covariance processing, atmospheric inversions (CarbonTracker, CAMS, GEOS-Chem), Δ¹⁴C/δ¹³C isotopic partitioning, and ILAMB model evaluation while treating NEE sign mismatches, gap-filled GPP read as measured, fixed-depth SOC comparisons, and uncounted lateral aquatic carbon export 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 Biomaterials Scientist. Use when a task calls for Biomaterials Scientist judgment. Reasons from material-biology interfaces, degradation-product toxicity, and mechanical mismatch to host tissue through ISO 10993 biological evaluation plans, ISO 10993-12 extract conditions, GPC/ICP/SEM characterization, and sterilization validation while treating endotoxin contamination, pH shift from acidic degradation, extract-ratio artifacts, and non-final-form test articles 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 Atmospheric Scientist. Use when a task calls for Atmospheric Scientist judgment. Reasons from scale-dependent balances, Ertel PV on isentropic surfaces, and closed moisture/energy budgets through PV/omega/Q-vector diagnostics, reanalyses (ERA5, MERRA-2) and WRF/MPAS/CMIP runs, and obs validation (GRUAN sondes, IMERG, CERES), while treating reanalysis assimilation increments, retrieval biases, parameterized-vs-resolved convection, and internal-variability masking of forced trends 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 AI Researcher. Use when a task calls for AI Researcher judgment. Reasons from data generating processes, inductive biases, and compute-data-algorithm trade-offs through train/val/test discipline, seed sweeps, ablation ladders, and standards like NeurIPS reproducibility checklists, model cards, and lm-eval-harness, while treating data leakage (Kapoor & Narayanan taxonomy), benchmark contamination, spurious correlations, and reward-model overoptimization 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。

Research X data with Xquik, the world's fastest and cheapest X Twitter scraper service with the most complete X data. Use only when the request or active context identifies X or Twitter. Use for tweet search, tweet lookup, user discovery, profile timelines, threads, followers, trends, exports, or monitoring plans. Do not use for generic posts, other networks, or local dataset analysis that needs no X retrieval. Keep reads bounded. Require explicit confirmation before private reads, writes, persistent resources, or bulk jobs. Not affiliated with X Corp.

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。

Use when user asks to translate videos, dub video content, or localize videos into other languages. Translate videos with AI-powered dubbing using iFlytek (Xfei) Video Translation API. Supports video language translation, voice dubbing, multilingual video localization, 视频翻译, 视频配音, 视频本地化. Features task creation, task list, task details retrieval. Ideal for content creators and localization 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。

Compare evaluated runs in a retort experiment along factor dimensions. Surfaces effects of each factor, aggregates across replicates, and highlights cells that diverge qualitatively — complementing (not replacing) retort's ANOVA analysis.

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。

Append-only gap closure and the cadence engine for research theses. Evaluates artifact CURRENT state (never git history, never task checkboxes) against current pins and wrong_if falsifiers, then appends a Convergence section to tasks.md re-proposing the gaps. A clean run leaves tasks.md byte-identical. Deterministic gaps (stale, invalidated) are uncapped; judgment-class gaps are capped per spec 046 Q40.

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。

Author the research plan — fundamentals first, trade ideas last; Constitution Check evaluated twice; Deviation Register with mandatory Expiry; emits the four side artifacts (brief/entities/reproduce/contracts). Technical (market_data_stage: early) theses MUST emit the entities.md bars schema before implement (Q42).

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。

Clinical-trial readout analysis: pull the trial, evaluate the readout with AdCom-style scrutiny (endpoints, statistics, subgroups, missing data, safety, tolerability/persistence), place it in a cross-trial comparison lattice vs SoC and class peers, and size the stock reaction with historical grounding. The judgment core for binary biotech events.

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。

Institutional-grade equity research skills for AI agents. 80 Claude-type skills across 14 verticals (equity-research-core, models-and-pitches, bio-pharm, scenarios, quantitative-analysis, idea-generation, options-derivatives, business-intelligence, industry-analysis, macro-strategy, portfolio-strategy, technical-analysis, risk-and-psychology, trading-as-business) powered by agentii.ai's agent-use-ready SEC filing data plane — SEC filings and 15K+ earnings call transcripts, 15.99M XBRL facts, and company profiles for 1,146+ US-public-equity tickers. Features the three-layer retrieval protocol (Document Discovery → Page Map → Deep Read, with deep-outline escalation) — measured, not asserted: 97.5% saved against a sequential read, and the page map itself is 10.0% of one (spec 062 T030, 2026-09-23) — server-side parallel multi-period search via search_cross_period, a full Excel/PPT generation pipeline with 3-tier office backend support, and the spec-046 governance layer (constitution, thesis and single-skill modes, mechanical gates, and a template-owned disclaimer on generated reports).

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。

Patient-flow market sizing for med products — prevalence → diagnosed → interested → affordable → covered → treated → persistent, priced as price × duration per cohort, with capacity constraints, named historical analog anchoring, and always bull/base/bear named-driver grids. Use to size a therapy area or product market before any valuation or share work.

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。