177 skills · sorted by relevance
Build normal traffic profiles for industrial control networks and identify unusual connections, polling intervals, and protocol operations. Use machine learning and process records to investigate deviations and produce anomaly reports with severity levels.
Investigate malware that survives OS reinstallation by checking UEFI firmware, boot files, and Secure Boot settings. Use chipsec and supporting tools to compare trusted baselines, identify bootkit indicators, and document findings.
Analyze access logs from AWS S3, GCS, and Azure Blob Storage to spot unusual downloads, unfamiliar IP addresses, and request spikes. Compare activity with normal patterns and prioritize findings for suspected data theft investigations.
Analyze NetFlow v9 and IPFIX records, which summarize network connections, to spot traffic spikes, scanning, unusual outbound transfers, and periodic connections. Compare traffic against normal patterns and produce a prioritized findings report.
Investigate hidden attacker activity using focused hypotheses, indicators of compromise such as malicious domains, and statistical checks on logs. Rank investigation priorities, generate indicator sweep targets, and flag unusual behavior for human review.
Add or update Handsontable performance tests for rendering and user interactions. The skill guides trace recording, measurement boundaries, baseline comparisons, and reports that help reviewers identify performance changes in pull requests.
Monitor daily AWS spending by service and account, compare it with historical baselines, and send Slack alerts with cost differences and likely causes. Known planned increases are excluded, and cloud resources remain unchanged.
Examine writing for familiar AI patterns, statistical features, Pangram detection scores, and hidden character signatures. Compare drafts against known human writing to interpret the evidence, with explanations and limitations rather than a definitive authorship claim.
Save a page's SEO settings and compare them with later versions to detect changes after updates or deployments. Review differences in titles, headings, indexing controls, structured data, and available performance metrics, with severity levels and stored comparison history.
Design, review, and implement Gantt charts for project schedules. Assess source data, task dependencies, and resource needs, then define suitable views, editing behavior, rendering tools, accessibility, exports, and tests—or recommend a clearer alternative.
Guides LoopX benchmark experiments through benchmark-toolkit, covering run management, comparisons under matched conditions, and analysis of agent action histories. It helps explain outcomes and efficiency while tracking source versions, score validity, and evidence boundaries.
Establish repeatable performance measurements for MSBuild and .NET builds before optimizing. Compare clean-output, changed-source, and unchanged builds, record timing ranges, and assess output layouts, dependency references, and parallel build options against the measured baseline.
Analyze MSBuild binary logs and timing summaries to find why builds are slow. Identify costly reference resolution, code analysis, file copying, restore work, or serialized tasks, then prioritize improvements by their share of total build time.
Organize experiment goals, settings, metrics, and progress in existing Obsidian notes. Keep temporary logs in daily notes, promote stable findings into result notes, and connect evidence with interpretation and decisions about the next step.
Production-ready reinforcement learning algorithms (PPO, SAC, DQN, TD3, DDPG, A2C) with scikit-learn-like API. Use for standard RL experiments, quick prototyping, and well-documented algorithm implementations. Best for single-agent RL with Gymnasium environments. For high-performance parallel training, multi-agent systems, or custom vectorized environments, use pufferlib instead.
The English summary is being prepared. Showing the original description.
LQF Machine Learning Expert Guide - Routed skill for ML/Statistical Modeling with Critical Discussion Mode. Triggers on: machine learning, modeling, prediction, training, classification, regression, clustering, deep learning, neural network, model evaluation, feature engineering, hyperparameter tuning, overfitting, underfitting, baseline, ablation study, critique my approach, review my model, is this a good idea, should I use, what's wrong with, evaluate my solution, challenge my assumptions, discuss my approach Engages in critical discussion with minimum 3 rounds of iterative refinement. Challenges both user proposals and own suggestions with fact-based critique. Demands evidence and baselines before accepting solutions.
The English summary is being prepared. Showing the original description.
Compare current benchmark results against historical baselines to spot performance regressions. Use as an explicit/manual helper for build-to-build degradation review, not for broad optimization strategy or low-level profiling ownership.
The English summary is being prepared. Showing the original description.
Autonomous NeMo-RL research agent workflow for directed hypothesis testing and open-ended discovery. Guides agents through the full experiment lifecycle: understanding recipes and environments, wiring RL or NeMo-gym runs, launching reproducible baselines and iterations, analyzing results, preserving human oversight, and using git plus TSV logs as the research ledger. Do NOT use for: bug fixes, code review, documentation, refactoring, dependency updates, or single-file changes.
The English summary is being prepared. Showing the original description.
Set up CIAgent regression testing for the AI agent in this repo — write a runner, record golden baselines, generate a test spec, and verify it. Use when the user asks to add tests, evals, or regression testing for their AI agent, or to set up CIAgent.
The English summary is being prepared. Showing the original description.
Verify and release Three.js browser games: playtest QA, automated bot playtests, mobile and responsive checks, production builds, static-hosting base paths, debug gating, bundle review, screenshots, visual regression baselines, canvas-pixel inspection with measured metrics, and release risk reports.
The English summary is being prepared. Showing the original description.
Choose and audit startup metrics using Croll and Yoskovitz's "Lean Analytics". Use when the user mentions "what metrics should we track", "KPIs", "north star metric", "One Metric That Matters (OMTM)", "vanity metrics", "analytics dashboard", "DAU/MAU", "churn benchmark", or "measure product-market fit". Also trigger when choosing metrics for a startup or feature, auditing a dashboard for vanity metrics, setting metric targets and baselines, or instrumenting a product by business model and stage. Covers good-vs-vanity metrics, the One Metric That Matters, metrics by business model, the five startup stages, and benchmarks. For the build-measure-learn loop, see lean-startup. For fixing activation and retention, see improve-retention.
The English summary is being prepared. Showing the original description.
Use when configuring, running, or fixing PHPStan static analysis in WordPress projects (plugins/themes/sites): phpstan.neon setup, baselines, WordPress-specific typing, and handling third-party plugin classes.
The English summary is being prepared. Showing the original description.
Runs servers and corporate systems — patching, configuration baselines, change control, capacity, and the routine that prevents incidents. Use this to establish a patching cadence, standardize server configuration, plan a maintenance window, decide change control for infrastructure, or clean up systems that have drifted from any known state.
The English summary is being prepared. Showing the original description.
Matches a user’s biomedical research direction, disease problem, study aim, data modality, and resource constraints to the most relevant recent algorithms and method papers. Always search real recent algorithm literature first, prioritize the last 12 months, expand to 1–3 years only when needed, and add canonical baselines only when necessary. Every formal algorithm recommendation must include the verified primary method paper, plus published downstream papers that actually cite/use the algorithm when such papers are found, with DOI when available. Never fabricate papers, algorithm names, authors, journals, years, DOI, PMID, links, or benchmark claims. If no directly verified algorithm paper is found, say so explicitly.
The English summary is being prepared. Showing the original description.