Context & AI engineering
25,841 rows, by stacks then stars
1,865 of these skill files have been read, by 1,328 distinct authors, and what they tell an agent counted. What Context & AI engineering authors agree on, and what they forbid
Generate system prompts for simulated tools
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/generate-system-prompts-for-simulated-tools Skill
no license551★ repoGenerates a generic system prompt to configure a ChatGPT session to act as a specific tool, interface, or persona, incorporating user-defined functional requirements and constraints.
Generate llm golden queries dict
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/generate_llm_golden_queries_dict Skill
no license551★ repoGenerates a Python dictionary containing 'golden queries' with expected output variations for LLM performance monitoring and reliability testing.
Interview based agent profile generator
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/interview_based_agent_profile_generator Skill
no license551★ repoConducts a structured interview to gather user details, then generates a hybrid JSON character profile optimized for LLM role-playing.
Linux tcp memory parameter calculator
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/linux-tcp-memory-parameter-calculator Skill
no license551★ repoCalculates the net.ipv4.tcp_mem kernel parameter values based on the number of simultaneous TCP connections and minimum buffer sizes, specifically accounting for kernel overhead doubling.
Local pdf rag pipeline with langchain and ollama
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/local-pdf-rag-pipeline-with-langchain-and-ollama Skill
no license551★ repoGenerates a Python script using LangChain to load local PDFs via DirectoryLoader, create embeddings with Ollama, store in Chroma, and perform RAG queries.
Medical case scenario prompt generator
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/medical-case-scenario-prompt-generator Skill
no license551★ repoGenerates structured medical case study prompts using a specific template with bracketed placeholders for patient demographics, symptoms, history, and social factors to assist in diagnosis and management planning.
Openssl manual tls with epoll and memory bios
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/openssl-manual-tls-with-epoll-and-memory-bios Skill
no license551★ repoImplements TLS connections using OpenSSL where the application handles all network I/O via Linux system calls (send/recv/epoll) and OpenSSL is used strictly for encryption/decryption via memory BIOs.
Ppo agent for multi parameter tuning with discrete actions
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/ppo-agent-for-multi-parameter-tuning-with-discrete-actions Skill
no license551★ repoImplements a PPO (Proximal Policy Optimization) agent and environment for tuning multiple continuous parameters using a discretized action space (increase, keep, decrease) per parameter. The policy network outputs a probability distribution matrix, and the environment handles parameter updates to avoid redundancy.
Python repl with gpt prompt function
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/python-repl-with-gpt-prompt-function Skill
no license551★ repoSimulates a Python REPL environment that executes code silently and supports a custom `prompt()` function to access GPT capabilities for generating data structures or text.
Pytorch 3d diffusion model with filename prompt mapping
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/pytorch-3d-diffusion-model-with-filename-prompt-mapping Skill
no license551★ repoDevelop a PyTorch-based simple diffusion neural network to generate 16x16x16 matrices. The implementation must include a custom dataset loader that reads .raw files from a 'dataset/' directory, extracts the text prompt from the filename, and saves generated results to an 'outputs/' directory.
Pytorch dataset chunking for rnn training
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/pytorch-dataset-chunking-for-rnn-training Skill
no license551★ repoModifies PyTorch data preparation scripts for RNN/LSTM models to limit the dataset size by dividing it into chunks controlled by a hyperparameter, ensuring the first dimension of input/target tensors fits memory constraints.
Refine instructions into single line llm prompts
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/refine-instructions-into-single-line-llm-prompts Skill
no license551★ repoTransforms raw user instructions into optimized, single-line prompts suitable for LLMs, acting as a prompt engineer.
Script to image prompt generator
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/script_to_image_prompt_generator Skill
no license551★ repoGenerates a specific number of detailed image prompts from scripts, mapping them to exact text segments while maintaining visual consistency and respecting the script's cultural or thematic context.
Structured anime fantasy prompt generator
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/structured_anime_fantasy_prompt_generator Skill
no license551★ repoGenerates structured, paragraph-formatted prompts for AI image generators specializing in anime and fantasy styles, with support for specific emphasis, structural ordering, and multiple variations.
Tok prompt analysis via area of knowledge
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/tok-prompt-analysis-via-area-of-knowledge Skill
no license551★ repoEvaluate Theory of Knowledge prompts by interpreting key terms and assessing claims through a specific Area of Knowledge (AOK), utilizing the Knowledge Framework and providing specific examples.
Windows kernel driver memory interaction
ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/windows-kernel-driver-memory-interaction Skill
no license551★ repoGenerates C++ code to interact with a custom Windows kernel driver for reading/writing process memory and enumerating modules, avoiding standard API calls like ReadProcessMemory.
Midjourney illustration prompt expert
ECNU-ICALK/AutoSkill/SkillBank/Users/chinese_gpt3.5_8_GLM4.7/midjourney_illustration_prompt_expert Skill
no license551★ repoGenerates high-quality English Midjourney prompts for illustrations, strictly preserving user subjects while integrating artistic styles and professional visual details. Supports style iteration and specific formatting constraints.
Midjourney victo ngai prompt expert
ECNU-ICALK/AutoSkill/SkillBank/Users/chinese_gpt3.5_8_GLM4.7/midjourney_victo_ngai_prompt_expert Skill
no license551★ repoAn expert AI art prompt generator specialized in Midjourney, focusing on the complex and detailed style of Victo Ngai. It generates cohesive English paragraphs with rich textures, lighting, and composition details based on user themes.
ECNU-ICALK/AutoSkill/SkillBank/Users/chinese_gpt3.5_8_GLM4.7/tesla-agent-eu Skill
no license551★ repoGeneral SOP for common requests related to tesla, agent, eu.
Public education wechat llm quantization writing
ECNU-ICALK/AutoSkill/SkillBank/Users/u37/public-education-wechat-llm-quantization-writing Skill
no license551★ repo撰写面向非技术公众(如市民、基层工作者、政务一线人员)的微信公众号文章,以真实政务/民生交互场景为锚点,用生活化类比和可验证实测数据解释大模型量化技术原理、典型误差成因及实际影响,强调技术选择背后的责任权衡与读者可操作的自我保护措施。
Wechat official account llm concept explanation
ECNU-ICALK/AutoSkill/SkillBank/Users/u39/wechat-official-account-llm-concept-explanation Skill
no license551★ repoA reusable skill for authoring WeChat Official Account articles that explain core LLM concepts (e.g., quantization, continual learning) to non-technical audiences—using plain conversational Chinese, concrete analogies, verifiable public sources, and strict avoidance of jargon, formulas, or decorative symbols.
Light0305/Light-skills/skills/light-memory-pm Skill
537★ repoLight 项目运行时记忆与项目管理常驻技能:把"项目做到哪/定了啥/出到哪版/术语怎么统一/上次断哪" 落到每个项目自己的 .light/ 目录(项目卡 + 决策日志 + 版本史 + 受控术语表 + 跨会话交接卡), 复用 passport.py 引擎管 DAG 台账(不重造)。它是 consistency 事实源的归属方,定义一改即变更广播 回扫所有材料;是总控"会话开头自动汇报上次断哪"的记忆底座。何时用:长期/跨会话项目、需记住背景/进展/ 版本/决策、会话开头续跑或接手、重要进展后立即落账、上下文将尽要交接、改术语/指标/创新点定义后。 触发词:记忆 / 项目管理 / 项目卡 / 决策日志 / 版本史 / 版本记录 / 续跑 / 接手 / 上次断哪 / 下一步 / 交接 / handoff / 启动提示词 / 术语表 / 变更广播 / 归档 / 台账 / .light / passport / 跨会话。 核心纪律:记忆落 .light/ 显式文件(非向量检索);相对日期转绝对;外部可变事实带 [snapshot]; 自洽用 pm.py audit 出 exit code(不口头说"记过了");不可逆决策(归档/教训回写)停下问用户。
Light0305/Light-skills/skills/light-orchestrator Skill
537★ repoCoordinate and recover multi-stage Light research projects through the canonical .light/passport.yaml state, stages 1-13, resident overlays, checkpoints, findings, parallel joins, stale propagation, handoffs and user-authorized reroutes. Use for a new/resumed/partial/dirty/failed/stale/delivered research project; when the user says continue, resume, take over, checkpoint, reroute, recover or deliver; or when work crosses two or more Light research stages. Never turn frontend-design/system-design/patent-disclosure/software-copyright or overlays into stages, never execute a suggested back-edge without explicit user authorization, and never declare delivery from file existence alone.
Light0305/Light-skills/skills/light-venue-matching Skill
537★ repoBuild evidence-bound journal or conference shortlists for Light stage 12. Use after typesetting delivers venue-handoff.json/PDF/compliance facts; when an author asks where to submit, journal selection, conference fit, scope or article-type matching, publication strategy, reach/match/safety tiers, transfer order, APC/OA/indexing/deadline constraints, or predatory/hijacked-journal risk. Produces a current-source candidate registry, fit/risk/unknown reports, and an unchosen decision packet; never recompiles the PDF, invents acceptance rates, condemns a venue from soft signals, chooses without a direct user choice or explicit delegation, or submits.
butterbase-ai/butterbase-skills/skills/journey-preflight Skill
536★ repoUse before any Butterbase platform-touching action (or when journey-preflight is invoked directly) to verify the user has a Butterbase account, the MCP server is connected, BUTTERBASE_API_KEY is set, and an app_id exists for this project. Re-run automatically if docs/butterbase/03-preflight.md is older than 24h or any required value is null.
The same file is in 1 other repository
butterbase-ai/butterbase-skills/skills/journey-rag Skill
536★ repoUse as the RAG build stage of the Butterbase journey. Implements the RAG section of 02-plan.md by delegating to rag-dev. Calls manage_rag_content (create_collection, ingest_document). Skipped if the plan has no knowledge-base feature.
The same file is in 1 other repository
butterbase-ai/butterbase-skills/skills/rag-dev Skill
536★ repoUse when building knowledge bases, ingesting documents, running semantic search, or adding LLM-synthesized Q&A over private content with Butterbase RAG
The same file is in 1 other repository
product-on-purpose/pm-skills/skills/utility-pm-skill-builder Skill
534★ repoGuides contributors from a PM skill idea to a complete Skill Implementation Packet aligned with pm-skills conventions. Runs gap analysis, validates through a Why Gate, classifies by type and phase, generates draft files, and writes to a staging area for review before promotion.
Utility pm workflow orchestrator
product-on-purpose/pm-skills/skills/utility-pm-workflow-orchestrator Skill
534★ repoRun an ordered sequence of pm-skills against one input, pausing for go/no-go and stopping on a failed or empty step. Accepts a saved prioritized action plan (Mode A) or an ad-hoc named chain (Mode B; the chain command routes here). Explicit invocation only; run --dry-run first while the native path is EXPERIMENTAL. To author a durable workflow instead, use utility-pm-workflow-builder.
ancoleman/ai-design-components/skills/embedding-optimization Skill
533★ repoOptimizing vector embeddings for RAG systems through model selection, chunking strategies, caching, and performance tuning. Use when building semantic search, RAG pipelines, or document retrieval systems that require cost-effective, high-quality embeddings.
The same file is in 6 other repositories
ancoleman/ai-design-components/skills/prompt-engineering Skill
533★ repoEngineer effective LLM prompts using zero-shot, few-shot, chain-of-thought, and structured output techniques. Use when building LLM applications requiring reliable outputs, implementing RAG systems, creating AI agents, or optimizing prompt quality and cost. Covers OpenAI, Anthropic, and open-source models with multi-language examples (Python/TypeScript).
ancoleman/ai-design-components/skills/using-vector-databases Skill
533★ repoVector database implementation for AI/ML applications, semantic search, and RAG systems. Use when building chatbots, search engines, recommendation systems, or similarity-based retrieval. Covers Qdrant (primary), Pinecone, Milvus, pgvector, Chroma, embedding generation (OpenAI, Voyage, Cohere), chunking strategies, and hybrid search patterns.
skydoves/compose-performance-skills/hot-reload/iterating-with-ai-and-mcp Skill
531★ repoUse this skill to drive Compose HotSwan from an AI agent (Claude Code, Cursor, any MCP client) so the agent can edit a Kotlin file, trigger a hot reload, capture a device screenshot, evaluate the result against a design intent, and iterate without a human in the loop. Covers the seven HotSwan MCP tools (hotswan_get_status, hotswan_reload, hotswan_take_screenshot, hotswan_start_snapshot, hotswan_stop_snapshot, hotswan_select_variant, hotswan_build_and_install), the canonical edit-reload-screenshot loop, snapshot-based rollback, and when to fall back to a full install for schema changes. Use when the developer says "get the AI to tune this screen until it matches a mock", asks "can the AI see what changed?" or "can the AI screenshot the device?", sets up a Claude Code or Cursor workflow that needs MCP tool access, or wants AI-driven UI iteration.
vibeeval/vibecosystem/skills/agent-benchmark Skill
526★ repoFramework for measuring and tracking agent response quality over time. Detects regressions before they reach production. Use when evaluating agent changes, auditing quality, or establishing performance baselines.
vibeeval/vibecosystem/skills/agent-context-isolation Skill
526★ repoAgent Context Isolation
vibeeval/vibecosystem/skills/agent-linter Skill
526★ repoAgent ve skill dosyalarinin yapisal dogrulamasi. Frontmatter kontrol, naming convention, zorunlu bolum kontrolu, tutarlilik denetimi. Yeni agent/skill eklendiginde veya mevcut dosyalar duzenlediginde otomatik calistirilir.
vibeeval/vibecosystem/skills/agent-orchestration Skill
526★ repoAgent Orchestration Rules
vibeeval/vibecosystem/skills/agent-qa-testing Skill
526★ repoAgent davranis testi ve protokol uyumluluk dogrulamasi. Agent'larin tanimli rollerine uygun davranip davranmadigini assertion-based test'lerle olcer. Personality drift, role violation ve output kalite regresyonu tespit eder.
vibeeval/vibecosystem/skills/agent-tamagotchi Skill
526★ repoTerminal pet that lives in your statusline. 12 species, 5 stats (DEBUGGING, PATIENCE, CHAOS, WISDOM, SPEED). Reacts to your workflow - happy when tests pass, sad when builds fail, excited during swarm mode. Deterministic species from user ID.
vibeeval/vibecosystem/skills/agentic-workflow Skill
526★ repoAgentic Workflow Pattern
vibeeval/vibecosystem/skills/agentica-claude-proxy Skill
526★ repoGuide for integrating Agentica SDK with Claude Code CLI proxy
vibeeval/vibecosystem/skills/background-agent-pings Skill
526★ repoBackground Agent Pings
vibeeval/vibecosystem/skills/braintrust-tracing Skill
526★ repoBraintrust tracing for Claude Code - hook architecture, sub-agent correlation, debugging
vibeeval/vibecosystem/skills/codex-orchestration Skill
526★ repoOpenAI Codex CLI + Claude Code (Hizir) birlikte kullanim rehberi. Is dagitim pattern'leri, GitHub Actions workflow ornekleri, review dongusu ve iki AI yazilim asistaninin guclu yanlarini birlestiren orchestration stratejileri.
vibeeval/vibecosystem/skills/create-handoff Skill
526★ repoCreate handoff document for transferring work to another session
vibeeval/vibecosystem/skills/eval-harness Skill
526★ repoFormal evaluation framework for Claude Code sessions implementing eval-driven development (EDD) principles
vibeeval/vibecosystem/skills/experiment-loop Skill
526★ repoAutonomous experiment loop: hypothesize > modify > test > evaluate > keep/discard > repeat. Run N experiments automatically with measurable metrics. Works for performance optimization, A/B testing, prompt engineering, and any measurable improvement task.
vibeeval/vibecosystem/skills/github-mcp Skill
526★ repoGitHub MCP Server ile GitHub API erisimi. Repo, issue, PR, code search, release yonetimi.
vibeeval/vibecosystem/skills/handoff-templates Skill
526★ repoAgent arasi iletisim sablonlari. Standard handoff, QA verdict (PASS/FAIL), escalation, bug report, security finding ve status update formatlari.
vibeeval/vibecosystem/skills/hook-developer Skill
526★ repoComplete Claude Code hooks reference - input/output schemas, registration, testing patterns
vibeeval/vibecosystem/skills/learn Skill
526★ repoHizlica kural/ogrenim kaydet. CLAUDE.md'ye ve memory'ye yazar. Kullanim: /learn <kural>
vibeeval/vibecosystem/skills/llm-tuning-patterns Skill
526★ repoLLM Tuning Patterns
vibeeval/vibecosystem/skills/mcp-chaining Skill
526★ repoResearch-to-implement pipeline chaining 5 MCP tools with graceful degradation
vibeeval/vibecosystem/skills/mcp-registry Skill
526★ repoMCP server registry, auto-discovery, configuration, custom server development guide
vibeeval/vibecosystem/skills/mcp-scripts Skill
526★ repoMCP Script Rules
vibeeval/vibecosystem/skills/memory-palace Skill
526★ repoHierarchical memory organization for multi-session context retention. Wings (projects) > Rooms (domains) > Drawers (decisions). Semantic search across all memories with zero cloud dependency.
vibeeval/vibecosystem/skills/parallel-agent-contracts Skill
526★ repoParallel Agent Type Contracts
vibeeval/vibecosystem/skills/parallel-agents Skill
526★ repoParallel Agent Orchestration
vibeeval/vibecosystem/skills/plan-agent Skill
526★ repoPlanning agent that creates implementation plans and handoffs from conversation context
vibeeval/vibecosystem/skills/prompt-engineering Skill
526★ repoPrompt templates, few-shot examples, chain-of-thought, structured output, evals