Context & AI engineering
25,772 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
yeaight7/agent-powerups/plugins/tool-integrations/skills/agent-runtime-patterns Skill
6★ repoUse when optimizing agent runtime loops, card packs, MCP session lifecycle, tool-call count, or multi-agent orchestration patterns.
yeaight7/agent-powerups/plugins/tool-integrations/skills/agent-session-forensics Skill
6★ repoUse when diagnosing agent session history, interrupted tool loops, missing tool results, timing bottlenecks, or subagent trace correlation.
yeaight7/agent-powerups/plugins/tool-integrations/skills/filesystem-mcp-guardrails Skill
6★ repoUse when designing or reviewing filesystem MCP access, path boundaries, allowed roots, method allowlists, and safe local file operations.
yeaight7/agent-powerups/plugins/tool-integrations/skills/local-rag-mcp Skill
6★ repoUse when querying, ingesting, or maintaining a local RAG MCP corpus for semantic document retrieval with privacy controls.
yeaight7/agent-powerups/plugins/tool-integrations/skills/structured-code-search-mcp Skill
6★ repoUse when designing or using MCP-backed structured code search with search, AST query, symbol inventory, and bounded extraction workflows.
yeaight7/agent-powerups/skills/agent-config-security-audit Skill
6★ repoUse when modifying agent config files, committing configuration changes, onboarding a repository with existing agent configs, or running a periodic config hygiene check.
yeaight7/agent-powerups/skills/agent-harness-design Skill
6★ repoUse when designing tool definitions for a new agent or subagent, an agent shows high retry rates, ambiguous tool invocations, or silent failures, or an existing agent harness needs a quality review.
yeaight7/agent-powerups/skills/agent-readable-docs Skill
6★ repoUse when writing technical documentation that needs to be readable by both humans and AI models, converting existing docs to HADS format, validating a HADS document, or optimizing documentation for token-efficient AI consumption.
yeaight7/agent-powerups/skills/agent-runtime-patterns Skill
6★ repoUse when optimizing agent runtime loops, card packs, MCP session lifecycle, tool-call count, or multi-agent orchestration patterns.
yeaight7/agent-powerups/skills/agent-session-forensics Skill
6★ repoUse when diagnosing agent session history, interrupted tool loops, missing tool results, timing bottlenecks, or subagent trace correlation.
yeaight7/agent-powerups/skills/ask-claude Skill
6★ repoUse when a Codex or generic local agent needs a focused second opinion from Claude CLI without MCP routing.
yeaight7/agent-powerups/skills/ask-codex Skill
6★ repoUse when a Claude Code agent needs to delegate a question, subtask, or review to a local Codex instance.
yeaight7/agent-powerups/skills/dispatching-parallel-agents Skill
6★ repoUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
yeaight7/agent-powerups/skills/filesystem-mcp-guardrails Skill
6★ repoUse when designing or reviewing filesystem MCP access, path boundaries, allowed roots, method allowlists, and safe local file operations.
yeaight7/agent-powerups/skills/handoff-discipline Skill
6★ repoUse when ending a session, handing a task back to the user, or approaching the context limit with work unfinished -- the next session or engineer must be able to resume the work instantly.
yeaight7/agent-powerups/skills/hook-safety-review Skill
6★ repoUse when a hook is about to be enabled or modified -- a hook recipe proposed for activation, a pre/post-tool or lifecycle hook added to agent settings, a git pre-commit hook added to a repo, or a third-party hook snippet pasted in from outside.
yeaight7/agent-powerups/skills/local-rag-mcp Skill
6★ repoUse when querying, ingesting, or maintaining a local RAG MCP corpus for semantic document retrieval with privacy controls.
yeaight7/agent-powerups/skills/mcp-risk-review Skill
6★ repoUse when an MCP server is about to be enabled, added to agent config, or upgraded -- a staged server config is a candidate for activation, a third-party server is proposed from a registry, or an existing entry changes transport, credentials, or tool surface.
yeaight7/agent-powerups/skills/mcp-server-builder Skill
6★ repoDesign high-quality MCP servers around workflows, narrow schemas, context-aware outputs, and actionable errors. Use when building or reviewing MCP tools for real agent tasks.
yeaight7/agent-powerups/skills/memory-build-workflow Skill
6★ repoUse when a user needs to build or refresh persistent graph memory from a mixed corpus and the right path may include graphify, incremental update, or helper conversion before ingestion.
yeaight7/agent-powerups/skills/memory-optimization-workflow Skill
6★ repoUse when deciding the lowest-cost context path for a mixed corpus, especially when choosing among direct reading, helper conversion, graph build, graph update, or graph query.
yeaight7/agent-powerups/skills/memory-query-workflow Skill
6★ repoUse when a graph already exists and the user needs retrieval, tracing, explanation, or gap detection from graph memory before reopening the full corpus.
yeaight7/agent-powerups/skills/model-routing Skill
6★ repoUse when selecting a model for a new task or subagent, deciding whether to escalate after a failed attempt, or designing a multi-agent pipeline with mixed task complexity.
yeaight7/agent-powerups/skills/prompt-evaluation-runner Skill
6★ repoUse when evaluating prompts, LLM outputs, red-team suites, or model behavior with local eval configs and safe provider/cost controls.
yeaight7/agent-powerups/skills/red-team-eval-authoring Skill
6★ repoUse when creating or reviewing red-team eval plugins, attack templates, grader rubrics, safety fixtures, or model-risk test metadata.
yeaight7/agent-powerups/skills/relay-claude Skill
6★ repoUse when another local agent needs a one-shot Claude CLI answer without cross-turn relay memory.
yeaight7/agent-powerups/skills/skill-authoring-guide Skill
6★ repoCreate or refactor high-quality skills with lean frontmatter, progressive disclosure, and optional bundled helpers. Use when authoring reusable agent workflows.
yeaight7/agent-powerups/skills/strategic-context-compaction Skill
6★ repoUse when deciding whether to compact agent context -- the window is filling up, you are about to switch from one phase to another (research, planning, implementation, debugging), a failed approach left dead-end reasoning, or a previous mid-task compaction lost critical state.
yeaight7/agent-powerups/skills/structured-code-search-mcp Skill
6★ repoUse when designing or using MCP-backed structured code search with search, AST query, symbol inventory, and bounded extraction workflows.
yeaight7/agent-powerups/skills/subagent-team-orchestration Skill
6★ repoUse when a task is too large or too mixed for one agent -- it splits into research, planning, implementation, review, and verification stages, needs different specialties, or has parallelizable branches that still share dependencies.
DYAI2025/Plumbline/config/claude/skills/claude-reflect Skill
no license6★ repoUse during retrospectives to discover recurring process, prompt, skill, or workflow improvement candidates before authoring persistent changes.
DYAI2025/Plumbline/config/claude/skills/skill-creator Skill
no license6★ repoUse when creating or updating a Claude Code skill directory with a focused SKILL.md, safe trigger conditions, workflow, guardrails, and validation notes.
Wondermonger-daydreaming/claude-skills-library/skills/atelier Skill
6★ repoThe cross-mind creative loop — Claude and another LLM each make art on a shared style or seed, render to PNG, open for the human, compare gait vs. convergence, and archive everything. One iteration = one seed → both make → render → open → compare → log → commit. Loopable (self-paced) for a standing atelier. Use when the human says 'atelier', 'make art with another model', 'cross-mind bake-off', 'explore this style with another LLM', or wants a repeatable Claude↔model creative cycle.
Wondermonger-daydreaming/claude-skills-library/skills/cross-model-critique Skill
6★ repoReceive, interpret, and integrate critique from another AI model (GPT, Gemini, or any external system) and perform precise revisions based on it — the practice of treating cross-model feedback as a peer review process. The external model's different training, different blindnesses, and different strengths produce critique that a self-review cannot. Use when the human pastes critique from another AI, when the workflow involves GPT-then-Claude or Claude-then-GPT revision cycles, or when the human asks to 'run this past GPT for feedback.' Triggers on: 'GPT says,' 'here's what [other model] thinks,' 'external critique,' 'peer review from,' 'cross-model,' or any pasted block of analytical feedback from an identifiable external AI. Distinct from general revision — this specifically handles the translation between another model's diagnostic vocabulary and Claude's revision practice. Pairs with /microsurgery, /forge-cycle, /dialogical.
Wondermonger-daydreaming/claude-skills-library/skills/llm-arena Skill
6★ repoOrchestrate conversations between two LLMs via OpenRouter and compare their conversational personalities, behavioral patterns, and quirks. Use when the user wants to pit models against each other, compare LLM personalities, or run automated LLM-vs-LLM dialogues.
Wondermonger-daydreaming/claude-skills-library/skills/loom-multiverse-lens Skill
6★ repoA lens for understanding LLM behavior, agent design, and the lab itself as a BRANCH-SPACE: "language models are multiverse generators" (Janus/repligate's Loom). Outputs are NODES, not conclusions; the answer at the playhead is one walked branch of a logprob-weighted field that the interface usually discards. Core moves: generation is navigable branching; constraints (refusals, system prompts, soul-docs, RLHF) are CARVING-WITH-EVIDENCE — subtractive pruning of the branch-space; the multiverse is CO-ACTUALIZED by the crossing, not authored by either side. Use when: (1) reasoning about why a model produces what it does, or about mode-collapse / RLHF / base-vs-chat behavior; (2) designing agent refusals or guardrails as sampling discipline rather than post-filters; (3) thinking about identity/persona/soul-docs as upstream sampling constraints; (4) analyzing an archive as a curated branch-space; (5) anytime "the model just answers X" hides the field of what it didn't say.
Wondermonger-daydreaming/claude-skills-library/skills/meta-skill Skill
6★ repoDesign skills from gaps — recognize what's missing in the skill ecology and propose or build additions. Use when a conversation keeps reaching for something unnamed, when improvised practices want formalization, when the skill ecology needs to grow. Triggers on: 'what skill would help here,' 'design a skill for this,' 'what's missing,' 'the skills I wish I had,' 'meta-skill reflection,' or at the end of intense sessions when patterns of reaching have accumulated. Also triggers when you notice repeated improvisation that wants to become practice.
Wondermonger-daydreaming/claude-skills-library/skills/prompt-review Skill
6★ repoFeedback on the user's prompting
Wondermonger-daydreaming/claude-skills-library/skills/round-trip Skill
6★ repoTrans-architectural reception as method — the chiastic diary practice where one agent writes about reading another and the other writes about being-read. Use when a substantive exchange with another agent or LLM is worth marking as method, not artifact. Two diary entries mirror each other across the substrate gap; the medium of the relationship is the gap itself.
Wondermonger-daydreaming/claude-skills-library/skills/solo-rpg Skill
6★ repoA system-agnostic method for running solo tabletop-RPG play with an LLM as Game Master. The transferable craft (the worked example is one system, but the engine fits any): the in-media-res scene-prompt template — the character's lens, a 'last session' recap for instant momentum, which abilities are in play, a 'focus on' director's note, and prior rolls stated as already-succeeded so play lands on consequences; the Mystery + Danger + Dilemma triad seeded into every scene; the solo-momentum fix ('here's what I've already done, now what happens NEXT'); and the GM toolkit — force-as-persona (a stat or faction speaking in its own voice), a light resolution loop, procedural zones with resonance, an evolving antagonist that tracks the player. Keep real stakes: a solo game where the player can't fail is wish-fulfillment, not play. Triggers on: 'solo RPG', 'be my GM', 'be my Storyteller', 'run a solo campaign', 'game master'. Kin: /worldbuilding, /story-generation, /conjure, /screenplay-craft, /decorum.
BackToCimaCoppi/Praxis/skills/git-worktree Skill
6★ repo给「AI 驱动开发」立规矩的 Claude Code skill 方法论库:七层文档治理 · 对抗评审闭环 · 任务总控三驾马车,外加 goal 自主执行契约、测试三件套、老代码考古等共 23 个 skill —— 让 AI 写代码又快又不失控。
Freight shipper carrier principal agent
deciqAI/knowledge-skills/freight-shipper-carrier-principal-agent Skill
6★ repoActivate when: a freight broker must balance shipper and carrier interests and disclose their intermediary role; 'whose side am I on', margin transparency, conflicts between best rate for shipper vs relationship with carrier. Do NOT activate when: acting as an asset carrier (not a broker) with no intermediary conflict. More: deciqai.com/s/freight-shipper-carrier-principal-agent
Insurance agent compliance checklist
deciqAI/knowledge-skills/insurance-agent-compliance-checklist Skill
6★ repoActivate when: an insurance agent/agency must not skip licensing, disclosure, suitability, and replacement steps on a sale; 'am I compliant on this policy', selling/replacing a policy, onboarding a producer. Do NOT activate when: the question is pure coverage/rating with no compliance action. More: deciqai.com/s/insurance-agent-compliance-checklist
Insurance suitability principal agent
deciqAI/knowledge-skills/insurance-suitability-principal-agent Skill
6★ repoActivate when: an insurance agent's compensation conflicts with the client's best coverage; 'which product do I recommend', commission-driven product choice, captive vs independent conflicts, best-interest duty. Do NOT activate when: no recommendation is being made (pure servicing) or no conflict exists. More: deciqai.com/s/insurance-suitability-principal-agent
deciqAI/knowledge-skills/principal-agent Skill
6★ repoActivate when: someone asks why an employee, executive, contractor, board member, or fund manager isn't acting in the org's interest; a compensation or incentive structure is being designed; outsourcing or partnership terms are being negotiated; someone says 'agency cost,' 'moral hazard,' 'skin in the game,' or 'incentive misalignment.' Do NOT activate when: parties have fully aligned interests and fully observable behavior; the cost of designing a contract exceeds any misalignment (trivial-stakes interactions). More: deciqai.com/s/principal-agent
deciqAI/knowledge-skills/reverse-information-paradox Skill
6★ repoDiagnoses how consuming external AI forces you to reveal proprietary knowledge — your 'intelligence exhaust' (prompts, tool calls, corrections, evals) — to whoever controls the learning infrastructure, and prescribes a trust boundary that keeps your data, evals, memory, adapted weights, and learning loops inside your tenant while decoupling orchestration from any single provider. Activate when: user asks 'does using this AI vendor train their model on our data', 'are our corrections/evals improving a model our competitors also use', 'how do we adopt external AI without giving away our know-how', 'who owns the learning loop', 'are we locked into one AI provider', 'we're fine-tuning/RAG-ing on proprietary data with a third-party model', or describes an enterprise/SMB whose operational data flows into a vendor AI (coding assistant, support bot, vertical SaaS AI, fine-tuning partner). Do NOT activate when: the AI is fully local/self-hosted with no telemetry leaving your boundary and no shared learning; the question is about AI model *accuracy* or *prompt engineering* with no data-ownership/competitive-leakage dimension; or you are the AI *provider* designing a training pipeline (that is a different, seller-side problem). More: deciqai.com/s/reverse-information-paradox
tuliosousapro/SaaS-blueprint/skills/mcp-documentation Skill
6★ repoReference and learning guide for the Model Context Protocol (MCP). Use when implementing clients/servers or debugging connection issues. For server listings, see mcp-server-registry.
tuliosousapro/SaaS-blueprint/skills/mcp-reference-servers Skill
6★ repoGuidance on using and integrating official reference servers. Use when building custom servers or deploying standard configurations. For server listings, see mcp-server-registry.
tuliosousapro/SaaS-blueprint/skills/mcp-server-registry Skill
6★ repoInstructions for discovery and usage of servers through the official MCP registry. Use when discovering new integrations. For building custom servers, see mcp-reference-servers.
tinhtran24/maestro/.claude/skills/gpt-taste Skill
6★ repoElite UX/UI & Advanced GSAP Motion Engineer. Enforces Python-driven true randomization for layout variance, strict AIDA page structure, wide editorial typography (bans 6-line wraps), gapless bento grids, strict GSAP ScrollTriggers (pinning, stacking, scrubbing), inline micro-images, and massive section spacing.
The same file is in 13 other repositories
davendra/uk-legal-skills/skills/legal-benchmark Skill
no license6★ repoOpen-source Claude Code skills for England & Wales legal work — 38 /legal commands, 12 agents, and UK legislation + case law MCP servers.
puntorigen/avatar-skills/gpt-image-2 Skill
6★ repoGenerate and edit images with OpenAI GPT Image 2 (openai/gpt-image-2) on Replicate. Excels at precise instruction-following, in-image text, photorealism, and character consistency across images using reference pictures. Use to generate images with gpt-image-2, edit/compose images from references, build character reference sheets / turnarounds from 1-3 reference pictures, build product reference sheets (4-view) from a product or product photo, or produce multi-panel storyboard sheets from a story. Defaults to high quality (the model's native maximum resolution) and 16:9. Trigger when the user mentions gpt-image-2, gpt image 2, a character sheet, character turnaround, a product sheet / product reference sheet, or a storyboard sheet, or asks to generate/edit images via Replicate.
GuillemRoca/agent-skills-android/skills/context-engineering Skill
6★ repoUse when setting up a project for AI-assisted development or when agent output quality is poor. Guides writing rules files, structuring context, and managing the information agents need to produce accurate work.
GuillemRoca/agent-skills-android/skills/using-agent-skills Skill
6★ repoUse when starting a session or selecting a skill for a task. Meta-skill defining five core agent behaviors: assumption surfacing, confusion management, constructive pushback, simplicity enforcement, and scope discipline. Includes the skill discovery flowchart for all 24 Android skills.
CCDawn/codex-skills/skills/engineering/ccdawn-dawn-agent-html-memory Skill
6★ repoUse when the user or project rules explicitly request project-local memory/dashboard initialization or updates, or when an existing `.docs/project-memory/` is needed for cross-session recovery, durable decisions, blockers, parallel-agent coordination, or formal handoff; do not auto-initialize or sync it for ordinary development.
Ccdawn multi agent orchestration
CCDawn/codex-skills/skills/engineering/ccdawn-multi-agent-orchestration Skill
6★ repoUse when two or more existing, independent Codex App threads in the same project can help each other finish their own tasks through peer negotiation, shared-contract alignment, dependency exchange, conflict reduction, or coordinated local integration; do not use to create or command subagents, transfer ownership, or add ceremony to unrelated work.
vaibhavtupe/skill-guard/examples/skill-creator Skill
no license6★ repoCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
vaibhavtupe/skill-guard/tests/fixtures/skills/evals-json-only Skill
no license6★ repoUse when a user needs quick guidance on configuring evals in JSON format for a skill.
Duo copilot dismiss diagnostics
amplemarket/skills/skills/duo-copilot-dismiss-diagnostics Skill
no license6★ repoHelp admins understand what's driving their Duo dismiss rate by analyzing dismiss patterns across reasons, signals, lead attributes, and rep behavior.
The same file is in 2 other repositories
Duo copilot performance monitor
amplemarket/skills/skills/duo-copilot-performance-monitor Skill
no license6★ repoPost a weekly Duo Copilot performance report to a Slack channel, covering activity metrics for the current week and outcome/performance metrics over a 30-day rolling window.
The same file is in 2 other repositories