Context & AI engineering
25,745 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
multiplex-ai/muggle-ai-teams/skills/ai-seo Skill
2★ repoWhen the user wants to optimize content for AI search engines, get cited by LLMs, or appear in AI-generated answers. Also use when the user mentions 'AI SEO,' 'AEO,' 'GEO,' 'LLMO,' 'answer engine optimization,' 'generative engine optimization,' 'LLM optimization,' 'AI Overviews,' 'optimize for ChatGPT,' 'optimize for Perplexity,' 'AI citations,' 'AI visibility,' 'zero-click search,' 'how do I show up in AI answers,' 'LLM mentions,' or 'optimize for Claude/Gemini.' Use this whenever someone wants their content to be cited or surfaced by AI assistants and AI search engines. For traditional technical and on-page SEO audits, see seo-audit. For structured data implementation, see schema-markup.
The same file is in 2 other repositories
multiplex-ai/muggle-ai-teams/skills/autonomous-loops Skill
2★ repoPatterns and architectures for autonomous Claude Code loops — from simple sequential pipelines to RFC-driven multi-agent DAG systems.
The same file is in 1 other repository
multiplex-ai/muggle-ai-teams/skills/claude-api Skill
2★ repoAnthropic Claude API patterns for Python and TypeScript. Covers Messages API, streaming, tool use, vision, extended thinking, batches, prompt caching, and Claude Agent SDK. Use when building applications with the Claude API or Anthropic SDKs.
The same file is in 2 other repositories
multiplex-ai/muggle-ai-teams/skills/claude-devfleet Skill
2★ repoOrchestrate multi-agent coding tasks via Claude DevFleet — plan projects, dispatch parallel agents in isolated worktrees, monitor progress, and read structured reports.
The same file is in 6 other repositories
multiplex-ai/muggle-ai-teams/skills/continuous-agent-loop Skill
2★ repoPatterns for continuous autonomous agent loops with quality gates, evals, and recovery controls.
The same file is in 2 other repositories
multiplex-ai/muggle-ai-teams/skills/cost-aware-llm-pipeline Skill
2★ repoCost optimization patterns for LLM API usage — model routing by task complexity, budget tracking, retry logic, and prompt caching.
The same file is in 9 other repositories
multiplex-ai/muggle-ai-teams/skills/data-scraper-agent Skill
2★ repoBuild a fully automated AI-powered data collection agent for any public source — job boards, prices, news, GitHub, sports, anything. Scrapes on a schedule, enriches data with a free LLM (Gemini Flash), stores results in Notion/Sheets/Supabase, and learns from user feedback. Runs 100% free on GitHub Actions. Use when the user wants to monitor, collect, or track any public data automatically.
The same file is in 8 other repositories
multiplex-ai/muggle-ai-teams/skills/dmux-workflows Skill
2★ repoMulti-agent orchestration using dmux (tmux pane manager for AI agents). Patterns for parallel agent workflows across Claude Code, Codex, OpenCode, and other harnesses. Use when running multiple agent sessions in parallel or coordinating multi-agent development workflows.
multiplex-ai/muggle-ai-teams/skills/enterprise-agent-ops Skill
2★ repoOperate long-lived agent workloads with observability, security boundaries, and lifecycle management.
The same file is in 7 other repositories
multiplex-ai/muggle-ai-teams/skills/eval-harness Skill
2★ repoFormal evaluation framework for Claude Code sessions implementing eval-driven development (EDD) principles
The same file is in 9 other repositories
multiplex-ai/muggle-ai-teams/skills/mcp-server-patterns Skill
2★ repoBuild MCP servers with Node/TypeScript SDK — tools, resources, prompts, Zod validation, stdio vs Streamable HTTP. Use Context7 or official MCP docs for latest API.
The same file is in 4 other repositories
multiplex-ai/muggle-ai-teams/skills/prompt-optimizer Skill
2★ repoAnalyze raw prompts, identify intent and gaps, match ECC components (skills/commands/agents/hooks), and output a ready-to-paste optimized prompt. Advisory role only — never executes the task itself. TRIGGER when: user says "optimize prompt", "improve my prompt", "how to write a prompt for", "help me prompt", "rewrite this prompt", or explicitly asks to enhance prompt quality. Also triggers on Chinese equivalents: "优化prompt", "改进prompt", "怎么写prompt", "帮我优化这个指令". DO NOT TRIGGER when: user wants the task executed directly, or says "just do it" / "直接做". DO NOT TRIGGER when user says "优化代码", "优化性能", "optimize performance", "optimize this code" — those are refactoring/performance tasks, not prompt optimization.
The same file is in 3 other repositories
multiplex-ai/muggle-ai-teams/skills/ralphinho-rfc-pipeline Skill
2★ repoRFC-driven multi-agent DAG execution pattern with quality gates, merge queues, and work unit orchestration.
The same file is in 7 other repositories
multiplex-ai/muggle-ai-teams/skills/regex-vs-llm-structured-text Skill
2★ repoDecision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for low-confidence edge cases.
The same file is in 7 other repositories
multiplex-ai/muggle-ai-teams/skills/security-scan Skill
2★ repoScan your Claude Code configuration (.claude/ directory) for security vulnerabilities, misconfigurations, and injection risks using AgentShield. Checks CLAUDE.md, settings.json, MCP servers, hooks, and agent definitions.
The same file is in 9 other repositories
multiplex-ai/muggle-ai-teams/skills/skill-stocktake Skill
2★ repoUse when auditing Claude skills and commands for quality. Supports Quick Scan (changed skills only) and Full Stocktake modes with sequential subagent batch evaluation.
The same file is in 2 other repositories
ViktorsBaikers/DevRites/pack/.claude/skills/rite-handoff Skill
no license2★ repoUser-invoked handoff writer: sync chat-only context into `.devrites/` and write a fresh-agent handoff.
ViktorsBaikers/DevRites/pack/generated/claude/skills/rite-handoff Skill
no license2★ repoUser-invoked handoff writer: sync chat-only context into `.devrites/` and write a fresh-agent handoff.
ViktorsBaikers/DevRites/pack/generated/codex/skills/rite-handoff Skill
no license2★ repoUser-invoked handoff writer: sync chat-only context into `.devrites/` and write a fresh-agent handoff.
kohoj/skills/codex-collab Skill
2★ repoBackground dual-agent collaboration with Codex CLI. Sends tasks to Codex via `codex exec` in the background, captures JSONL output, and integrates results without blocking Claude Code. No tmux, no Terminal windows. Use when asked to "codex collab", "dual agent", "pair with codex", "codex tmux", "collaborate with codex", "let codex help", or "spiral coding".
kohoj/skills/hermes-agent-dissect Skill
2★ repoNavigate and extract reusable design patterns from the Hermes Agent source code for application in our project. It covers ten subsystems: bootstrap/CLI, conversation engine, prompt/context, tool system, gateway/messaging, skills, terminal backends, memory/persistence, RL training, and integrations (ACP/MCP/plugins). Use this skill whenever a user wants to study, reference, or adapt architectural patterns from Hermes Agent. Also use it when the user asks questions such as "how does Hermes do this?", "learn from Hermes Agent", "dissect Hermes", "reference Hermes source", or requests implementation details for any Hermes Agent feature. This skill is the primary entry point: it routes to the appropriate reference file by topic, then reads the source code and returns extractable patterns. Even when the question appears simple (for example, "how does Hermes implement X?"), use this skill because the reference files provide precise file paths, reading order, and architectural context that substantially improve answer quality.
citedy/skills/skills/code-review-agent-team Skill
2★ repoParallel multi-agent code review using Agent Teams with 4 specialized reviewers. Spawns a coordinated team of security, performance, test coverage, and code quality agents. Teammates can share findings with each other for cross-domain insights. Produces unified report with severity-ranked findings saved to /output/. Requires Claude Code with Agent Teams support (TeamCreate, TaskCreate, SendMessage). Use when asked to: - "review code", "code review", "check my changes", "review PR" - "security review", "performance review", "test coverage check" - "full code review", "multi-agent review", "parallel review" Examples: - `/code-review-team` -- review uncommitted changes (git diff) - `/code-review-team 42` -- review PR #42 - `/code-review-team --scope=security` -- security-only review of uncommitted changes - `/code-review-team 42 --scope=performance,tests` -- specific reviewers for PR
The same file is in 2 other repositories
citedy/skills/skills/prompt-analyzer Skill
2★ repoAnalyze prompts for constraint complexity, audit failure risks, and generate optimized rewrites for Claude and GPT. Based on "How LLMs Follow Instructions" (Rocchetti & Ferrara, 2026) constraint taxonomy research. Use when reviewing prompt files, optimizing prompt bases, or auditing instruction quality. Trigger phrases — "analyze prompt", "prompt audit", "optimize prompts", "constraint analysis", "prompt review", "check prompts".
citedy/skills/skills/skill-quality-eval Skill
2★ repoValidate and audit quality of AI agent slash commands (.claude/commands/*.md) — checks YAML frontmatter, description length, jargon detection, directive language, and duplicate detection. Use when creating, editing, or auditing project commands and skills. Run proactively after adding new commands.
The same file is in 2 other repositories
citedy/skills/skills/spawning-plan Skill
2★ repoSpawning Plan. Use when user wants to spawn agents, create a team, or coordinate multiple agents. Automatically gathers context, asks team topology questions, outputs clean TEAM PLAN markdown, and gets user approval. 3 steps: context gathering -> questions -> present plan. Requires Claude Code with Agent Teams support.
The same file is in 2 other repositories
hajibabaie/combinatorial-optimization-skills/skills/particle-swarm-optimization Skill
2★ repoWhen the user wants to design, implement, or tune particle swarm optimization, covering velocity and position updates, inertia weight, constriction, swarm topologies, and discrete adaptations such as random-key and binary PSO. Also use when the user mentions "particle swarm," "PSO," "inertia weight," "velocity update," "swarm topology," or "binary PSO," or when a gradient-free continuous box-constrained problem needs a population method. For the often stronger vector-difference alternative, see differential-evolution; for mapping particles to permutations, see decoder-based-representations.
aj604/toolshed/plugins/doc-lifecycle/skills/writing-docs Skill
2★ repoUse when writing or editing a documentation file whose job is to track the repo — README, runbook, CLAUDE.md/AGENTS.md, reference (including the claim-style docs/reference/architecture.md that bootstrapping-docs prescribes), or agent-facing context pack — including converting human or marketing docs into dense agent-facing form, or when tempted to add example output, install steps, or "why" prose. The one door for these repo-tracking docs, across human and agent readers (token bloat, context rot). Out of scope: tutorials, narrative architecture/conceptual overviews and design rationale, and decision records (ADRs) — narrative by design, not line-by-line claims about the repo (growing-docs creates those when demand justifies it). To create a baseline doc set from scratch for an undocumented repo, start with bootstrapping-docs; it routes here for each doc.
matejformanek/postgres-claude/.claude/skills/memory-contexts Skill
2★ repoAllocate memory in PostgreSQL backend C — pick the right MemoryContext and use palloc / palloc0 / pstrdup / psprintf correctly. Covers CurrentMemoryContext / TopMemoryContext / per-query / per-tuple / ExecutorState context choice, MemoryContextSwitchTo discipline, the OOM-throws-ereport contract (no NULL checks), pfree vs MemoryContextReset vs MemoryContextDelete, the AllocSet vs Slab vs Generation vs Bump context-type cheat sheet, and leak-scoping in long-running backends. Use whenever a PG patch or extension calls palloc / palloc0 / MemoryContextAlloc, creates or switches a MemoryContext, picks AllocSet vs Slab vs Generation vs Bump, or debugs a context-shaped leak. Skip for plain malloc / free / jemalloc / mimalloc / tcmalloc, JVM / Go / .NET GC tuning, Rust Box / Rc / Arc / lifetimes, shared_buffers / work_mem production tuning, valgrind / heaptrack on non-PG programs, and C++ smart pointers.
matejformanek/postgres-claude/.claude/skills/memory-keeping Skill
2★ repoClose out a pg-claude session — sync progress/STATE.md, progress/coverage.md, progress/files-examined.md, and append a sessions/ log entry whenever a session produced durable output (a new knowledge/idioms or knowledge/subsystems doc, a [verified-by-code] fact, a discovered gotcha, a file-by-file deep read, or a locked decision from pg-claude-plan.md §14). Use proactively when the user says "wrap up", "close out the session", "sync memory files", "record this gotcha", "we're done for the day", or "log this for next time" — and whenever durable output was just produced. Skip for PG MemoryContext / palloc / pfree internals questions (use memory-contexts), LangChain / LangGraph / LlamaIndex agent memory and vector stores, application memory leaks / valgrind / heaptrack, conversational chat history persistence, finding old Claude sessions (use find-session), and ChatGPT / Claude.ai conversation export.
matejformanek/postgres-claude/.claude/skills/meta-commit-style Skill
2★ repoWrite a commit message inside the postgres-claude META repo (knowledge corpus, skill / agent / command edits, planning artifacts, session logs, infra changes) — covers the meta style: `ft(scope):` / `hf(scope):` / `docs(scope):` prefix, a wrapped body, an optional `Plan:` trailer, and the global `Co-authored-by: Claude Opus 4.7 (1M context) <[email protected]>` footer (lowercase trailer per real log — 45 lowercase vs 11 uppercase in last 50 commits). Distinct from the upstream PG commit-message-style (which forbids Co-authored-by and uses bare-imperative titles). Use whenever drafting a commit inside `postgres-claude/` (the meta repo) — corpus PR, skill rewrite, session log, planning doc, cloud-routine recipe, etc. Skip for upstream PG patches in `dev/` (use commit-message-style), Conventional Commits style (feat: / fix: / chore: across most non-PG projects), Linux-kernel Signed-off-by commits, Angular / Vue / React commit style, and the generic git-commit message question on any other project.
matejformanek/postgres-claude/.claude/skills/pg-claude Skill
2★ repoRoute any PostgreSQL backend-hacking task through the pg-claude meta repo's skill / slash-command / knowledge map — master index cataloguing every pg-claude skill, every slash command, and the knowledge/ corpus layout (subsystems, idioms, data-structures). Use proactively as the first-touch routing skill whenever the task is PG backend work: building from source, debugging a backend, writing C in src/backend, adding a GUC / index AM / table AM / access method, hacking the executor or planner, adding a SQL keyword, preparing a pgsql-hackers patch, reviewing a CommitFest entry, or any "where does X live in PG" navigation question. Skip for user-level SQL writing and tuning, autovacuum / shared_buffers / work_mem production DBA tuning, psycopg / pg8000 / node-postgres / pgx / JDBC / SQLAlchemy / Prisma / Django ORM client libraries, comparing PG distros (Aurora / Cloud SQL / Crunchy / Supabase / Neon), and SQL standard / ANSI compatibility questions.
danielrosehill/Claude-Israel-Agent-Skills-Plugin/skills/discover-israel-skills Skill
2★ repoUse when the user wants to browse, discover, or install third-party Claude Code agent skills focused on Israel (tax/accounting, government services, healthcare pharmacies, rail, cinema, post tracking, legal research, security compliance, communication, etc.) — skills authored by people other than Daniel, indexed here for easy installation alongside this plugin. Reads a curated list from `data/third-party-skills.yaml` (monorepo skill collections from github.com/skills-il plus standalone skill repos), shows each entry with its source repo / subpath / blurb, confirms the user's selection, and installs the chosen skill(s) into `~/.claude/skills/<name>/` via `git clone` or sparse-checkout. Does not install anything without explicit confirmation. Also surfaces the awesome-list reference (alexpolonsky/awesome-agent-skills-israel) for broader discovery. Trigger phrases - "what other israel skills are there", "install israeli agent skills", "find me a skill for [israel rail / clalit / maccabi pharmacy / cinema / israel post / tax / accounting]", "discover israel skills", "browse israel agent skills", "third-party israel skills".
danielrosehill/Claude-Israel-Agent-Skills-Plugin/skills/install-companion-plugins Skill
2★ repoUse when the user wants to install or review other Claude Code plugins that complement this Israel-Agent-Skills plugin. Reads a curated shortlist from `data/companion-plugins.yaml`, shows each plugin with its topic and blurb, confirms the user's selection, and runs the `claude plugins marketplace add` + `claude plugins install` commands. Does not install anything without explicit confirmation. The list of available companion plugins is maintained in the YAML file — do not enumerate them in the README or other user-facing docs; direct users to invoke the skill instead. Trigger phrases - "install related israel plugins", "what other plugins go with israel-agent-skills", "install companion plugins", "show me related plugins", "companion plugin install".
danielrosehill/Claude-Israel-Agent-Skills-Plugin/skills/list-skills Skill
2★ repoCategorized map of every skill in the israel-agent-skills plugin, grouped by area (Healthcare & Medication, Emergency Preparedness, Travel & Transit, Government Services, Finance, Media, Connectivity, Meta/Tooling). Use when the user asks "what skills are in this plugin", "is there a skill for X in Israel", "which medication / shelter / municipality skill should I use", or when you (Claude) need a quick categorical map without reading every individual SKILL.md. Also lists which skills are orchestrators (chain other skills) vs. direct lookups, so you pick the right entry point.
danielrosehill/Claude-Israel-Agent-Skills-Plugin/skills/miklatim-lookup Skill
2★ repoUse when the user wants to find a public bomb shelter (miklat tziburi, מקלט ציבורי) near a location in Israel — by address, neighbourhood, coordinates, or current position — with address, type, capacity, accessibility, and Google Maps / Waze directions. Primary backend is the Miklat-MCP server (streamable HTTP, `https://mcp.jlmshelters.com/mcp`), which currently supports Jerusalem (198 public shelters). If the MCP server is unavailable or the user is asking about a city not yet supported, fall back to the Jerusalem Municipality's official list at `https://www.jerusalem.muni.il/en/residents/security/spaces/list/` (for Jerusalem only). Trigger phrases - "find a shelter near me", "nearest miklat", "closest bomb shelter in Jerusalem", "miklat tziburi", "shelters in [neighbourhood]", "shelter capacity", "accessible shelter near", "how do I get to the shelter", "public shelters list Jerusalem".
ebarti/skills/ai-context-engineering Skill
2★ repoFoundational context engineering theory and practice for LLM applications. Covers the 5 levels of context (zero/linear/goal-oriented/role-based/semantic blueprint), semantic role labeling (SRL) for visualizing structured prompts, and the layered (scope → investigation → action) analysis pattern. Use this skill when: - Choosing between prompt engineering and context engineering for a task - Designing context that includes role, goals, and structured intent - Building a meeting / document / interview analysis pipeline - Visualizing the semantic structure of a prompt with SRL - Deciding when to upgrade a prompt's "context level"
ebarti/skills/ai-finetuning Skill
2★ repoPractical knowledge for finetuning foundation models. Covers when to finetune (vs prompt engineering or RAG), memory bottlenecks (backpropagation, quantization, numerical representations), parameter-efficient finetuning techniques (PEFT, LoRA, adapters), model merging strategies (summing, layer stacking, concatenation), and finetuning tactics (frameworks, hyperparameters). Use this skill when: - Deciding whether to finetune (vs prompt engineering or RAG) - Estimating memory requirements for finetuning - Implementing LoRA or other PEFT techniques - Merging multiple finetuned models - Choosing finetuning hyperparameters or frameworks - Quantizing models for inference or training
ebarti/skills/ai-multi-agent-mcp Skill
2★ repoBuilding multi-agent systems with a simplified inter-agent message envelope inspired by MCP-like structured messaging. This is not the official Model Context Protocol (MCP), which is a host/client/server protocol using JSON-RPC 2.0 with lifecycle initialization, capability negotiation, and server features such as tools/resources/prompts. Covers the book's pedagogical envelope format and transport choices, specialist agent design (Researcher / Writer + shared helpers), Orchestrator design with goal decomposition, and robustness via validation loops + a Validator agent. Use this skill when: - Designing inter-agent communication (simplified envelope schema, transport choice) - Building specialist agents (one role per agent, system prompt as identity) - Implementing an Orchestrator that routes between agents - Adding validation loops, retries, or a Validator agent - Hardening a multi-agent prototype against LLM/network/data failures
ebarti/skills/ai-production-architecture Skill
2★ repoPractical knowledge for architecting and operating AI applications in production. Covers the AI engineering architecture (context enhancement, guardrails, model router, gateway, caching, agent patterns), monitoring and observability (metrics, logs, traces, drift detection), pipeline orchestration, and user feedback systems (extracting conversational feedback, feedback design, biases, degenerate loops). Use this skill when: - Designing the architecture for a production AI application - Adding input/output guardrails - Setting up model routing or a gateway - Implementing caching (exact, semantic) - Building observability for an AI system - Designing user feedback collection
ebarti/skills/ai-prompt-engineering Skill
2★ repoPractical knowledge for writing, organizing, and defending prompts for foundation models. Covers in-context learning (zero-shot, few-shot), system vs user prompts, context efficiency, prompting best practices (clear instructions, examples, decomposition, chain-of-thought), prompt versioning, and defensive prompt engineering against jailbreaks and prompt injection attacks. Use this skill when: - Writing or improving prompts for an LLM application - Setting up prompt templates and versioning - Breaking down a complex task into prompts - Defending against prompt injection or jailbreaking - Designing system prompts and context structures - Evaluating prompt engineering tools
ebarti/skills/ai-rag-and-agents Skill
2★ repoPractical knowledge for building RAG (Retrieval-Augmented Generation) systems and AI agents. Covers RAG architecture and retrieval algorithms (term-based, embedding-based, hybrid), retrieval optimization (chunking, reranking, query rewriting), multimodal RAG, agent design with tools and planning, agent failure modes, and memory systems. Use this skill when: - Building a RAG system from scratch or improving an existing one - Choosing retrieval algorithms (BM25, dense vectors, hybrid) - Optimizing retrieval (chunking strategy, reranker selection) - Designing an AI agent (tools, planning, error correction) - Debugging agent failures - Implementing memory for conversational systems
ebarti/skills/ai-rag-defense Skill
2★ repoToken reduction, retrieval fidelity, and prompt-injection defense layers for the Context Engine. Covers a Summarizer agent for proactive context reduction, micro-context engineering (prompt design inside agents), high-fidelity RAG with source-metadata citations, layered untrusted-context defenses against prompt injection, and grounded-reasoning validation that prevents hallucination. Use this skill when: - Reducing token costs by adding a Summarizer agent - Designing system prompts inside specialist agents (micro-context) - Adding citations and source metadata to RAG outputs - Defending agents against prompt injection / data poisoning - Validating agent reliability across high-fidelity / summarization / grounded test cases - Preventing hallucination when retrieval returns no result
cleatai/agent-skills/skills/price-benchmark Skill
2★ repoBenchmark pricing for a government pursuit in CLEATUS from comparable award history. Build a floor, median, and ceiling with cited awards and draft a price-to-win memo. Run per pursuit before pricing review.
7ucg/baron-forge/plugins/baron-forge/skills/memory Skill
2★ repoPersistent cross-session memory for Baron. Use whenever Baron says remember/save/note this, asks what you remember, references a past decision/fact/config, or starts work that should continue from earlier state. Backed by a global markdown store (baron-mem) that loads automatically at session start and is editable via the /mem-save, /mem-load, /mem-set commands. Save: project state, decisions, conventions, endpoints, key offsets/field maps, gotchas, todos.
tkolleh/skills/smart-init Skill
2★ repoTrigger on: init, CLAUDE.md, AGENTS.md, setup repo, onboard, generate docs, project documentation for agents, bootstrap, new project setup. Evidence-guided AGENTS.md/CLAUDE.md generator that makes AI coding agent runs ~28% faster and ~16% cheaper by eliminating exploratory navigation.
ciaransaunders/Skills-Public/cowork-prompt-generator Skill
2★ repoGenerates optimized prompts for Claude Cowork (Anthropic's autonomous file-based agent mode). Use whenever the user wants to delegate any file-based, document, or multi-step task to Cowork — even if they don't say "Cowork" explicitly. Triggers include: "I want a prompt for Cowork", "do this in Cowork", "help me set up a Cowork task", "write a Cowork instruction", "can Cowork do X", or any request involving organising files, synthesising documents, bulk processing, or producing a professional output from local files. Always use this skill rather than writing a Cowork prompt ad hoc. Translates a vague goal into a structured, workspace-scoped Cowork prompt with clear outcomes, file scope, constraints, and output format.
ciaransaunders/Skills-Public/prompt-architect Skill
2★ repoCreate and optimize reusable prompt templates for AI assistants and autonomous agents. Use when user says "make a prompt for this task", "improve this prompt", "write a prompt template", "create a prompt for [task]", "optimize this prompt", or requests help designing instructions for LLMs or agents. Also trigger when the user asks about prompting best practices, how to prompt a specific model (Claude Fable 5, Claude Opus 4.8, GPT-5.5, Gemini 3.x), or wants to translate a prompt from one model to another. Incorporates vendor-specific prompt engineering guidance from Anthropic (Claude Fable 5/Mythos 5, Claude Opus 4.8), OpenAI (GPT-5.5), and Google (Gemini 3.5 Flash), plus research on the Prompting Inversion, Over-Specification Paradox, and the four-layer instruction stack for autonomous agents.
yacb2/aidex/skills/aidex-workflow Skill
2★ repoUse when the user wants to design or scaffold a one-shot multi-agent fan-out / decomposition orchestration — review code across N dimensions, migrate N call-sites in parallel, research N sources, or split an implementation into parallel agents with per-agent model/effort assignment — landing as a written `.context/workflows/` workflow-spec (goal + fan-out shape + per-agent model table + gate policy) before the Workflow runs. Fires on "design a workflow to review X across N angles", "fan out agents to migrate every call-site", "orchestrate a multi-agent review of X", "set up a workflow to research X from multiple sources", "decompose this into parallel agents with different models". Not for: executing a written multi-phase plan (aidex-plan-exec — there is a plan doc); repeat-until-a-check-passes automation (aidex-loop); planning multi-step work without orchestration (aidex-plan); ecosystem audits (aidex); project-state audits (aidex-audit).
yacb2/aidex/skills/aidex-worktree Skill
2★ repoUse when the user wants to create, run or tear down a git worktree with its own isolated environment (database, ports, containers), or when a project has no worktree setup yet and one needs bootstrapping by investigating the repo topology and verifying the compose stack can run twice — "set up a worktree for X", "create an isolated worktree", "spin up a second environment", "tear down the worktree", "how do we do worktrees on this project", "my worktrees are leaving Docker images/volumes behind", "worktree ports collide", first-time worktree setup. Also fires when aidex-plan/aidex-plan-exec/aidex-loop reach their Isolation step. Not for: planning multi-step work (aidex-plan); designing a loop (aidex-loop); a single-repo code-only checkout with no services (native EnterWorktree).
yacb2/aidex/skills/aidex Skill
2★ repoUse when the user wants to audit, organize, clean up, or health-check their Claude Code setup — a messy or inconsistent .context/, a bloated or stale MEMORY.md, broken symlinks in .claude/skills, unused or misplaced skills, dead CLAUDE.md links, plugins inflating idle context, or "my project opens heavy / wastes context". Also fires on "audit my project", "organize my ecosystem", "what skills do I have", "check my project's health", "I think I have broken symlinks", "my .context/ is a mess", "my new project has no .context / help me start with aidex", and the /aidex command. Not for: creating .context/ docs (aidex-conventions); project-state audits like UX or security (aidex-audit); backlog items (aidex-backlog).
Timmy6942025/opencode-builder-skill/skills/opencode-builder Skill
2★ repoUse this skill when the user is building extensions, plugins, integrations, or custom developer tools on top of OpenCode (opencode.ai). This includes: creating plugins with custom AI tools and lifecycle hooks, using the SDK for programmatic session control and multi-agent workflows, building model providers or MCP integrations, configuring opencode.json, and implementing advanced orchestration like checkpoint/resume, parallel execution, or adversarial review. Trigger for any development work that extends or automates OpenCode. Do NOT trigger for general usage questions, troubleshooting, UI theming, or tool comparisons.
Timmy6942025/opencode-builder-skill/skills/opencode-mcp Skill
2★ repoUse this skill when adding Model Context Protocol (MCP) servers to OpenCode, configuring local or remote MCP servers, setting up OAuth authentication for MCP, managing MCP server permissions per agent, or troubleshooting MCP connection issues. Covers local/remote server types, OAuth flows, tool management, overriding remote defaults, and common MCP server examples.
aksheyw/career-command-center-template/skills/update-memory Skill
2★ repoLog what worked (or didn't) to improve future applications. Argument: [company] [outcome] [what worked / what to change]
WhiteMuush/Your-Claude-DevOps-Teacher/skills/dispatching-parallel-agents Skill
no license2★ repoUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
WhiteMuush/Your-Claude-DevOps-Teacher/skills/docker-swarm-guide Skill
no license2★ repoOrchestration avec Docker Swarm incluant services, stacks, overlay networks, secrets, rolling updates et haute disponibilité. Se déclenche avec "Docker Swarm", "swarm mode", "docker service", "docker stack", "overlay network"
WhiteMuush/Your-Claude-DevOps-Teacher/skills/subagent-driven-development Skill
no license2★ repoUse when executing implementation plans with independent tasks in the current session
ariana-dot-dev/cloud-sandbox-vm-skills/skills/parallel-agents Skill
2★ repoBuild an "agent factory" — configure one cloud VM once, then disk-level fork it into a fleet of identical isolated machines so many agents run in parallel, each with its own filesystem, IPv4, and SSH. Flat per-second pricing makes a fleet ~10x cheaper than E2B/Modal (50 VMs ≈ $432/mo vs ~$4k). Use to fan work across many real machines instead of threads.
RamanaNani/agent-router/skills/hina Skill
2★ repoYour conversational personal assistant for Claude Code. Use when the user types /hina or talks to Hina by name. She remembers you across sessions (file-based memory), asks before acting when context is missing, routes the work to the best installed specialist via agent-router, does it, and gets better from your feedback. Runs in the main loop so she can pause and ask you questions.
RamanaNani/agent-router/skills/skill-finder Skill
2★ repoBrowse and discover Claude Code skills, agents, and plugins from the marketplace — ranked by rating — that are NOT yet installed. Use when the user is new and has few/no specialists installed, asks "what plugins/skills are available for X", "find me a skill for Y", "what should I install", "browse the marketplace", or when agent-router finds no installed tool that fits. Returns a ranked shortlist with real ratings, source links, and the exact install command.
xonevn-ai/xone-trading-skills/skills/edge-candidate-agent Skill
2★ repoGenerate and prioritize US equity long-side edge research tickets from EOD observations, then export pipeline-ready candidate specs for trade-strategy-pipeline Phase I. Use when users ask to turn hypotheses/anomalies into reproducible research tickets, convert validated ideas into `strategy.yaml` + `metadata.json`, or preflight-check interface compatibility (`edge-finder-candidate/v1`) before running pipeline backtests.
The same file is in 1 other repository