1.3K skills · sorted by relevance
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.
The English summary is being prepared. Showing the original description.
Cross-model benchmark for gstack skills. Runs the same prompt through Claude, GPT (via Codex CLI), and Gemini side-by-side — compares latency, tokens, cost, and optionally quality via LLM judge. Answers "which model is actually best for this skill?" with data instead of vibes. Separate from /benchmark, which measures web page performance. Use when: "benchmark models", "compare models", "which model is best for X", "cross-model comparison", "model shootout". (gstack) Voice triggers (speech-to-text aliases): "compare models", "model shootout", "which model is best".
The English summary is being prepared. Showing the original description.
When the user requests information about academic benchmarks, datasets, or research papers, particularly in machine learning, deep learning, or logical reasoning domains. This skill enables systematic research of academic benchmarks by searching web sources, downloading and analyzing arXiv papers, extracting key metadata (number of tasks, training availability, difficulty levels), and compiling comparative summaries. It triggers on requests involving dataset comparisons, benchmark analysis, or academic paper research for table creation.
The English summary is being prepared. Showing the original description.
This skill details how to conduct cloud security audits using Center for Internet Security benchmarks for AWS, Azure, and GCP. It covers interpreting CIS Foundations Benchmark controls, running automated assessments with tools like Prowler and ScoutSuite, remediating failed controls, and maintaining continuous compliance monitoring against CIS v5 for AWS, v4 for Azure, and v4 for GCP.
The English summary is being prepared. Showing the original description.
Helps build Go tests using a test-first workflow. Covers grouped test cases, HTTP response checks, performance benchmarks, fuzzing to explore generated inputs, and coverage reports showing which code was exercised by tests.
Turn scored competitor profiles into a report for strategic decisions. Organize market maps, comparison tables, unoccupied positions, threats, and prioritized recommendations so leaders can decide whom to compete with and how to differentiate.
Turn speedup requests into a measured optimization process with limits on time, cost, and variants. Compare each change against a baseline and correctness checks, then retain the fastest safe option with reproducible commands and measurements.
Measure and optimize claude-flow v3 startup, search, attention processing, and memory use. Compare benchmarks against performance targets, track changes over time, and identify regressions to guide further optimization.
Measure throughput, response times, and resource use for systems and groups of agents. Use baseline comparisons, load tests, and performance criteria to identify slowdowns, assess behavior under pressure, and summarize results with recommendations.
Evaluate and improve the policy instructions that guide an AI agent using scored coding tasks. Run a learning cycle to compare candidates, or use the default dry run to inspect tasks and estimated costs without model calls.
Measure and optimize claude-flow v3 startup, search, attention processing, and memory use. Compare benchmarks against performance targets, track changes over time, and identify regressions to guide further optimization.
Benchmark how CodeGraph affects an agent investigating a real codebase. Choose a version, language, repository, and execution mode, then compare cost, tool calls, file reads, and answer correctness with and without CodeGraph.
Design evaluation tasks that check whether agents actually complete work. Define requests, simulated user behavior, starting conditions, and verifiable success criteria, then plan difficulty levels, training-data leakage checks, and ongoing maintenance.
Plan and audit HypoGeniC workflows that use language models to propose patterns from labeled text. Check configurations, dataset splits, cost limits, and saved predictions while keeping candidate hypotheses separate from scientific confirmation.
Check release readiness inside the herdr repository by comparing changes since the base release with upcoming release notes and documentation. Review CLI descriptions, issue references, checks, and benchmarks, then report readiness.
Review findings, drafts, assessments, and reports left by different work desks in a shared folder. Use each desk’s journal for context, then summarize the current state and highlight items that need attention.
Add a CUDA/C++ GPU operation to sgl-kernel using AOT, meaning compilation before use. Follow the implementation through PyTorch registration, Python API exposure, SGLang integration, correctness tests, and benchmarks that account for cache reuse.
Review SGLang Cookbook pull requests for configuration errors, documentation mismatches, benchmark issues, and broken links. Get findings for each file and an overall recommendation on whether to approve, request changes, or block the contribution.
Explain portfolio performance by separating the effects of sector allocation, stock selection, market factors, and timing. Compare results with a benchmark, assess risk-adjusted returns, and prepare a report showing where excess returns came from.
Analyze social media post data to calculate engagement, click rates, and estimated returns on advertising spend. Compare results with supplied platform benchmarks, identify strong and weak posts, and generate recommendations for improving campaign performance.
Create, update, run, and debug shell-based performance scenarios for the but CLI. It separates repository setup from the operation being measured, uses Hyperfine for timing, and guides focused validation and inspection of test output.
Turn a broad robotics research direction into ranked ideas with concrete benchmarks and simulation plans. The workflow combines literature review, novelty checks, and external critique while accounting for hardware access, safety, evaluation metrics, and failure cases.
Build Go services and command-line tools with clear interfaces, explicit error handling, and concurrent tasks using goroutines and channels. Use pprof to investigate performance, then check behavior with benchmarks, structured test cases, fuzzing, and race detection.
Run concurrent Omnigent conversations with real hosts and runners while replacing the LLM with a zero-latency mock. Read the generated report to explain response-time distributions, throughput, and failures, helping assess Omnigent's own processing overhead.