「benchmark」の検索結果

Use when you want to benchmark and improve an existing repo skill through repeated evals, prompt mutations, and score tracking.

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 you want campaign performance reviewed against benchmarks, with a health score and prioritized optimization recommendations.

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。

Skill Optimization: Auto Research. Autonomously optimize any Claude Code skill by running it repeatedly, scoring outputs against binary evals, mutating the prompt, and keeping improvements. Based on Karpathy's autoresearch methodology. Use when: optimize this skill, improve this skill, run autoresearch on, make this skill better, self-improve skill, benchmark skill, eval my skill, run evals on. Outputs: an improved SKILL.md, a results log, and a changelog of every mutation tried.

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。

Create user-centered, accessible interface copy (microcopy) for digital products including buttons, labels, error messages, notifications, forms, onboarding, empty states, success messages, and help text. Use when writing or editing any text that appears in apps, websites, or software interfaces, designing conversational flows, establishing voice and tone guidelines, auditing product content for consistency and usability, reviewing UI strings, or improving existing interface copy. Applies UX writing best practices based on four quality standards — purposeful, concise, conversational, and clear. Includes accessibility guidelines, research-backed benchmarks (sentence length, comprehension rates, reading levels), expanded error patterns, tone adaptation frameworks, and comprehensive reference materials.

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。

Daily ad performance check-in across platforms. Pulls live metrics from Meta, Google, and LinkedIn via deterministic scripts, compares against benchmarks and previous period, flags overspend/underperformers/policy issues, and outputs a quick daily summary with action items. Use when "daily ad review", "how are my ads doing today", "ad check-in", "morning ad report", "daily ad summary", "check ad performance", "ad dashboard", "daily ads", or any request for a recurring or quick-glance ad performance review.

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。

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

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。

Run realistic enterprise-style IT scenarios before trusting an automation agent in production operations.

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 EnterpriseRAG-Bench to evaluate an enterprise RAG or knowledge-agent system against a realistic synthetic company corpus with answer, recall, and comparative scoring.

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。

Run Claude Code, Codex CLI, Gemini CLI, or OpenCode through bounded H100 post-training tasks and compare how well each agent improves a base LLM.

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。

Run structured prompt-injection attack and defense experiments against an LLM-integrated app before production by measuring attack success and testing detection or recovery pipelines.

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。

Run a repeatable evaluation suite for browser agents against static web task snapshots instead of judging them from demos or one-off tests.

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。

Run concurrent scripted conversations against a target agent to measure whether it stays on task, responds correctly, and holds up in repeatable test cases.

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。

Nature-style engineering manuscript reviewer skill for testing devices, robots, platforms, systems, algorithms embodied in physical systems and engineering methods against requirements, validation, benchmarking, uncertainty, robustness and 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。

Rates any SKILL.md on a 0-100 scale across 10 dimensions with a SHIP / REWORK / SCRAP verdict. Evaluates trigger precision, instruction clarity, output predictability, edge case coverage, anti-hallucination guardrails, developer experience, composability, open-source readiness, wow factor, and real-world utility. No flattery — calibrated against Anthropic's own skill-creator rubric. Use this skill whenever someone says "score this skill", "rate my skill", "is this skill good", "skill review", "skill audit", "roast my SKILL.md", "grade this", "will this skill work", "evaluate my skill", "how good is this", or pastes a SKILL.md and asks for feedback. Also trigger when comparing two skills, benchmarking a skill collection, or asking "what's wrong with this skill" — even without the word "score".

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。

A/B test CLAUDE.md instruction changes against eval benchmarks. Capture baselines, test variants, compare results.

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。

Run eval scenarios to benchmark Mycelium effectiveness. Execute tasks using reflexion loop, validate against success criteria, record metrics.

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 auditing the imp codebase for structural debt, dead code, god-objects/files, duplication, flag sprawl, or deciding whether a cleanup is worth shipping - "structure audit", "tech debt", "is this still used", "dead code", "refactor for clarity", "should we split this file", "remove this flag", "file size gate red", "alloc sites gate red". Do NOT use for build/test mechanics (building-and-testing), perf/kernel work (benchmark-cuda / sm120-cuda-expert), or output quality (check-degeneration).

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 keeping imp's docs and config examples coherent after a change - updating docs/internals/ARCHITECTURE.md / README / docs/GOAL.md / docs/MODELS.md / docs/roadmap.md / imp.conf.example / CHANGELOG / MISSION_JOURNAL, or "is this doc stale", "document this change", "the example config is out of date", "the README says X but the code does Y", "add a roadmap ledger row". Do NOT use for layer/frontmatter/lint questions (docs-layers), structural code audits (codebase-audit), measuring perf or refreshing the perf baseline (benchmark-cuda), or the agent's private memory (MEMORY.md).

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 imp outputs differ between two paths that should agree - chunk sizes, prefill vs decode, prefix-cache resend, two engines (imp vs llama.cpp vs HF), two builds or kernels - "output changes with chunk size", "first token differs", "logprobs moved", "which layer diverges", "who is right, imp or llama.cpp", "near-tie". Do NOT use for repetition loops or garbage output (check-degeneration), a new arch that loads wrong (add-model-arch), or perf (benchmark-cuda).

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 working on imp's quantization formats, loaders, or dequant paths - GGUF Q4_0...Q8_0/Q*_K/IQ4/MXFP4, NVFP4 two-level scaling (Modelopt vs compressed-tensors layouts), FP8 E4M3, StorageTier, decode cache, KV-cache dtypes (auto/fp16/fp8/int8/int4/nvfp4/mxfp4), "which quant should I use", "which KV dtype", scale-factor layout, dequant kernel wiring, imp-quantize / AWQ, judging quant quality (PPL). Do NOT use for writing/optimizing GEMM/GEMV kernels (sm120-cuda-expert) or measuring quant perf (benchmark-cuda).

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 writing, reviewing, or optimizing CUDA kernels targeting sm_120a (RTX 5090 / Consumer Blackwell, GB202) in the imp inference engine. Triggers on CUDA/PTX kernel code, shared-memory layout, bank conflicts, tensor-core MMA (mxf4nvf4, tf32, 3xTF32/3xFP16), GEMV/GEMM/attention/quantization/GDN-scan kernels, decode tok/s under expected, kernel emitting HMMA instead of mxf4nvf4, occupancy, register pressure, __launch_bounds__, spills, PDL. Pair with `benchmark-cuda` for measurement and `check-degeneration` after hot-path changes.

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 opening, merging, or releasing a PR for imp - branching off main, `gh pr create`, enabling auto-merge, writing the PR body or a CHANGELOG entry, cutting a tagged release (version bump + CHANGELOG + tag + GitHub release). Symptoms - "PR stuck BLOCKED", "my last commit didn't land on main", "which check is required", "how do I cut a release", "auto-merge merged too early", "check-release failed", "STALE.md blocks git pull". Do NOT use for build/test mechanics (building-and-testing) or perf measurement / baseline refresh (benchmark-cuda).

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。

Constructs fund-level performance reporting with IRR, MOIC, DPI, RVPI, PME, and vintage year benchmarking. Use when building fund reports, calculating performance metrics, or preparing LP reporting packages.

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。

Monitors secondary market pricing trends with discount/premium analysis, bid-ask spreads, and market-clearing dynamics. Use when analyzing secondary pricing, tracking market trends, or benchmarking transaction levels.

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。