All authors
Claude Skills by Amey-Thakur
github.com/Amey-Thakur1,001 skills16 installs1,473 views
- Csharp LinqUse LINQ with deferred execution understood, database translation respected, and loops chosen when they win. Use when writing C# queries over collections or ORMs, or debugging surprising LINQ behavior.Votes: 0GitHub stars: 7
- Dotnet AsyncWrite async C# that never blocks on tasks, flows cancellation everywhere, and uses ValueTask where it pays. Use when writing .NET async code or debugging thread-pool starvation and deadlocks.Votes: 0GitHub stars: 7
- Dotnet Dependency InjectionUse .NET's built-in DI with correct lifetimes, the options pattern, and constructor injection free of service-locator drift. Use when wiring .NET services or debugging captive-dependency and disposal bugs.Votes: 0GitHub stars: 7
- Dotnet Minimal ApisBuild ASP.NET Core minimal APIs with organized endpoint groups, filters, typed results, and a clear line to MVC. Use when structuring .NET HTTP services or deciding between minimal APIs and controllers.Votes: 0GitHub stars: 7
- Java CollectionsSelect Java collections by access pattern, use streams where they clarify, and default to immutability and records. Use when modeling data in Java or reviewing collection-heavy code.Votes: 0GitHub stars: 7
- Jvm Gc SelectionChoose and size a JVM garbage collector from latency goals and allocation behavior, and read GC logs before tuning flags. Use when configuring JVM services or diagnosing pause and memory pressure problems.Votes: 0GitHub stars: 7
- Jvm Memory ModelApply happens-before reasoning, volatile, and safe publication to write correct concurrent Java/Kotlin. Use when writing shared-state JVM code or diagnosing visibility and reordering bugs.Votes: 0GitHub stars: 7
- Jvm ProfilingProfile JVM services with async-profiler and JFR, separating CPU, allocation, and lock evidence, safely in production. Use when a JVM service is slow, hot, or memory-hungry and you need the cause.Votes: 0GitHub stars: 7
- Kotlin IdiomsWrite Kotlin that uses null safety, data classes, and coroutines as designed, with clean Java interop. Use when writing Kotlin or migrating Java code and habits to it.Votes: 0GitHub stars: 7
- Spring Boot DisciplineKeep Spring Boot apps explicit with constructor injection, typed configuration, test slices, and restrained magic. Use when building or reviewing Spring Boot services that must stay debuggable.Votes: 0GitHub stars: 7
- Chess ImprovementImprove at chess through tactical pattern training, endgame fundamentals, and engine-assisted analysis of your own losses. Use when your rating has plateaued and playing more games is not helping.Votes: 0GitHub stars: 7
- Curriculum SequencingOrder material so each piece is learnable when it arrives, with prerequisites satisfied and difficulty increasing gradually. Use when designing a course, onboarding path, or tutorial series.Votes: 0GitHub stars: 7
- Deliberate PracticeImprove at a skill through focused repetition at the edge of ability with immediate feedback, rather than through accumulated hours. Use when you have plateaued despite continued practice.Votes: 0GitHub stars: 7
- Explaining To BeginnersExplain something to someone with no background, without condescension and without jargon that quietly requires prior knowledge. Use when writing introductory material or onboarding someone new.Votes: 0GitHub stars: 7
- Feedback For LearningGive feedback that improves performance, timed and framed so it is used rather than defended against. Use when coaching, reviewing work, or mentoring.Votes: 0GitHub stars: 7
- Knowledge TransferMove understanding from one person to others so critical knowledge does not leave with them. Use before someone changes role or when one person is the only one who understands something.Votes: 0GitHub stars: 7
- Learning From FailureExtract the lesson from a mistake by examining the reasoning that produced it, rather than only the outcome. Use after something goes wrong, personally or as a team.Votes: 0GitHub stars: 7
- Mental Model BuildingBuild an internal model of how a system works so you can predict its behaviour rather than recalling procedures. Use when learning a complex system or when you can follow steps but cannot debug.Votes: 0GitHub stars: 7
- Self Directed LearningLearn something substantial without a course, by defining the goal, finding the right resources, and building feedback into the process. Use when you need a capability nobody is going to teach you.Votes: 0GitHub stars: 7
- Skill AssessmentJudge someone's ability, including your own, through demonstrated performance rather than confidence or credentials. Use when hiring, mentoring, or deciding what to learn next.Votes: 0GitHub stars: 7
- Spaced RepetitionSchedule review at increasing intervals so knowledge is retained with minimal total time. Use when you must remember material long-term rather than pass an immediate test.Votes: 0GitHub stars: 7
- Teaching Technical ConceptsExplain a technical idea by starting from what the learner knows, one concept at a time, with a concrete example before the abstraction. Use when teaching, mentoring, or writing an explanation.Votes: 0GitHub stars: 7
- Agent MemoryDesign memory for an AI agent so it recalls what matters and forgets the rest, without drowning in its own history. Use when adding persistence, context recall, or long-running state to an agent.Votes: 0GitHub stars: 7
- Agentic LoopsBuild reliable agent loops (plan, act, observe) with termination, error recovery, and progress guarantees. Use when building an LLM agent that takes multiple tool-using steps toward a goal.Votes: 0GitHub stars: 7
- Coding Agent WorkflowDirect and supervise a coding agent (Claude Code, Cursor, Antigravity) so it ships correct, verified work. Use when delegating engineering tasks to an AI agent and you want reliable results, not plausible-looking ones.Votes: 0GitHub stars: 7
- Context EngineeringAssemble what goes into an LLM's context on each call: selecting, ordering, and formatting the right information. Use when deciding what to put in a prompt or debugging why a model ignores or misuses provided information.Votes: 0GitHub stars: 7
- Context Window ManagementBudget tokens across system prompt, history, and retrieval, with compaction that keeps what matters. Use when building long-running LLM applications or debugging context overflow and mid-conversation amnesia.Votes: 0GitHub stars: 7
- Conversation DesignDesign multi-turn conversational systems with state tracking, memory injection, topic handling, and repair. Use when building chat assistants or fixing bots that forget, drift, or trap users.Votes: 0GitHub stars: 7
- Embeddings SelectionChoose embedding models and chunking by evaluating retrieval on your own corpus, not by leaderboard rank. Use when building semantic search or RAG and deciding how to embed and chunk.Votes: 0GitHub stars: 7
- Fine Tuning Vs PromptingDecide between prompting, retrieval, and fine-tuning with eval-first discipline and honest data requirements. Use when someone proposes fine-tuning or a prompt has hit its ceiling.Votes: 0GitHub stars: 7
- Goal Driven ExecutionPursue a high-level goal to completion autonomously: lock the done-condition, decompose, execute with verification, adapt, and know when to stop. Use when an agent is handed an outcome to achieve rather than a step to perform.Votes: 0GitHub stars: 7
- Llm Cost LatencyCut LLM cost and latency with caching, model tiering, prompt diet, batching, and streaming UX. Use when the inference bill or response time needs engineering down.Votes: 0GitHub stars: 7
- Llm Eval DesignBuild LLM evaluations from real failures with calibrated judges and regression gates that catch quality drift. Use when an LLM feature needs quality measurement or prompts change without anyone knowing what broke.Votes: 0GitHub stars: 7
- Llm GuardrailsLayer input filtering, output validation, injection defense, and human escalation around LLM features. Use when an LLM system faces untrusted input or its outputs carry real-world consequences.Votes: 0GitHub stars: 7
- Llm ObservabilityInstrument LLM applications with request tracing, token accounting, quality signals, and feedback loops. Use when running LLM features in production or debugging why quality or cost moved.Votes: 0GitHub stars: 7
- Mcp ServerDesign a Model Context Protocol server that exposes tools and data to AI agents safely and legibly. Use when building an MCP server or deciding what to expose to an agent.Votes: 0GitHub stars: 7
- Prompt CachingCut LLM cost and latency by caching stable prompt prefixes, structuring prompts so the cache actually hits. Use when the same system prompt, tools, or examples repeat across calls and cost or time-to-first-token matters.Votes: 0GitHub stars: 7
- Prompt EngineeringBuild prompts that get accurate, reliably-shaped output from any LLM, choosing the right technique for the task. Use when writing, improving, or debugging a prompt.Votes: 0GitHub stars: 7
- Rag PipelineDesign a retrieval-augmented generation pipeline that answers from sources and can be trusted. Use when building or debugging RAG: chunking, embedding, retrieval, context assembly, and grounding.Votes: 0GitHub stars: 7
- Retrieval RerankingImprove RAG answer quality by reranking retrieved candidates so the most relevant chunks reach the model. Use when a retrieval system returns roughly-right documents but the best ones are not on top, or answers miss available evidence.Votes: 0GitHub stars: 7
- Self ReflectionHave an agent critique and improve its own output before delivering, catching errors a first pass misses, without spiraling. Use when correctness matters and an agent should check its own work rather than ship the first draft.Votes: 0GitHub stars: 7
- Staying On TaskKeep an AI agent anchored to the actual goal through long or messy work. Use at the start of any multi-step task, and whenever work begins to drift, balloon, or stall.Votes: 0GitHub stars: 7
- Structured OutputGet valid structured data from LLMs with schema-constrained generation, boundary validation, and repair loops. Use when LLM output feeds code, databases, or downstream systems.Votes: 0GitHub stars: 7
- Tool Use DesignDesign LLM tools with descriptions that steer, granularity that composes, and error returns the model can act on. Use when building agent tool sets or debugging wrong-tool and wrong-argument failures.Votes: 0GitHub stars: 7
- Cross ValidationPick CV schemes that respect time and grouping, nest them for tuning, and read the variance, not just the mean. Use when data is too small for a single split or when validating tuning claims.Votes: 0GitHub stars: 7
- Drift MonitoringDetect input, prediction, and performance drift with reference windows and act through retrain or rollback triggers. Use when operating models in production or diagnosing gradual quality decay.Votes: 0GitHub stars: 7
- Experiment TrackingRecord every training run's code, data, config, and results so any model is reproducible and comparisons are honest. Use when setting up ML experiment infrastructure or untangling which run produced the prod model.Votes: 0GitHub stars: 7
- Feature EngineeringBuild features that are leakage-free, temporally correct, and reproducible between training and serving. Use when creating model inputs or auditing a suspiciously good offline score.Votes: 0GitHub stars: 7
- Hyperparameter TuningSearch hyperparameters with budgets, early stopping, and validation hygiene so gains are real, not overfit to the dev set. Use when tuning models or reviewing tuning claims.Votes: 0GitHub stars: 7
- Imbalanced DataHandle skewed classes with threshold moving, weighting, and PR-based evaluation instead of reflexive resampling. Use when the positive class is rare and accuracy looks deceptively high.Votes: 0GitHub stars: 7