agentsclimarketplace

Context & AI engineering

25,729 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

  • Llm council prompts

    Paldom/llm-council-skills/skills/llm-council-prompts Skill

    1★ repo

    Writes the prompts for each LLM council stage - forced-perspective advisor prompts, anonymized peer review with a parseable ranking contract, decisive chairman synthesis that may side with a dissenter, rebuttal rounds. Use for "write the council/advisor/chairman prompts", "peer review prompt", "synthesis prompt". Not for pipeline architecture, member selection, or aggregation rules.

  • Llm council when

    Paldom/llm-council-skills/skills/llm-council-when Skill

    1★ repo

    Decides whether an LLM council (multi-model deliberation/ensemble) beats a single strong model - task-type triage, compute-normalized evidence, cost-per-resolved-outcome, benchmarking both paths on your workload. Use for "should I use multiple models", "is an LLM council/ensemble/debate worth it", "single model vs council". Not for designing the pipeline, picking members, or cutting cost.

  • Security pass

    yoelgal/better-dev/skills/security-pass Skill

    1★ repo

    Use when a diff or work-item touches an untrusted-input, auth, money/PII, file-upload, external-fetch, or LLM/agent surface and needs a security verdict - as /review's Security channel when the host ships no /security-review, as /plan-grill's abuse-case reference, invoked directly on a work-item, or when deciding whether some tool/log/model output is safe to act on.

  • Uninstall

    yoelgal/better-dev/skills/uninstall Skill

    1★ repo

    Use when a person wants to remove better-dev - unwire a repo (drop the .better-dev/bin bridge and optionally the CLAUDE.md/AGENTS.md block) or remove the global per-host install. Never deletes your rules, overrides, or learnings unless you ask.

  • Worktree branching

    yoelgal/better-dev/skills/worktree-branching Skill

    1★ repo

    Use when a feature, fix, or chore work-item is about to start and needs its own isolated branch off the integration branch - before planning, diagnosis, or the autonomous loop runs - or when detecting whether such a worktree already exists. For removing one after merge, see this skill's teardown notes; for moving a half-finished item to a colleague or another machine, its handoff notes.

  • Impl validator

    chobizzy/llm-wiki/skills/impl-validator Skill

    1★ repo

    Validate whether an implementation matches its stated goal. Use this skill when a skill or agent wants a second opinion on its own output, when the user says "check this implementation", "validate what you did", "is this correct?", "review the output", or "did you do this right?". Also spawned automatically as a subagent by other skills (memory-bridge, daily-update) to self-check their outputs before presenting to the user. Returns a structured pass/warn/fail verdict with specific actionable issues.

    The same file is in 1 other repository

  • Llm wiki

    chobizzy/llm-wiki/skills/llm-wiki Skill

    1★ repo

    The foundational knowledge distillation pattern for building and maintaining an AI-powered Obsidian wiki. Based on Andrej Karpathy's LLM Wiki architecture. Use this skill whenever the user wants to understand the wiki pattern, set up a new knowledge base, or needs guidance on the three-layer architecture (raw sources → wiki → schema). Also use when discussing knowledge management strategy, wiki structure decisions, or how to organize distilled knowledge. This is the "theory" skill — other skills handle specific operations (ingesting, querying, linting).

  • Memory bridge

    chobizzy/llm-wiki/skills/memory-bridge Skill

    1★ repo

    Browse and compare wiki knowledge by which AI tool originally produced it. Use this skill when the user says "/memory-bridge", "browse codex memory", "what did codex know about X", "show me claude knowledge", "cross-tool memory", "what does hermes know that claude doesn't", "show me knowledge from <tool>", "compare my AI tool memories", or wants to explore knowledge gaps between tools. Works from any project. Diff mode ("what's different", "unique to codex", "gaps between tools") is the killer feature — it surfaces blind spots between tools that the user may not know exist.

  • Skill creator

    chobizzy/llm-wiki/skills/skill-creator Skill

    1★ repo

    Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, 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.

    The same file is in 66 other repositories

  • Wiki agent

    chobizzy/llm-wiki/skills/wiki-agent Skill

    1★ repo

    Query-driven targeted ingest from a specific AI agent's raw history. Use this skill when the user invokes /wiki-claude, /wiki-codex, /wiki-hermes, /wiki-openclaw, /wiki-copilot, /wiki-pi — with or without a search topic. Different from wiki-history-ingest (which bulk-ingests everything new): this skill finds sessions about a SPECIFIC TOPIC in a specific agent's history and ingests just those, then returns a synthesized answer immediately usable in the current session. Primary use case: you're working in agent A and want to pull in how you solved X in agent B's history. Cross-referencing, not archiving. Also trigger on: "what did I work on in codex about X", "search my claude sessions for Y", "pull in hermes knowledge about Z", "find that conversation where I did X in codex".

  • Wiki history ingest

    chobizzy/llm-wiki/skills/wiki-history-ingest Skill

    1★ repo

    Unified wiki-history-ingest entrypoint for conversation/session sources. Use this when the user says "/wiki-history-ingest claude", "/wiki-history-ingest hermes", or asks to ingest agent history without naming the underlying skill. This router dispatches to the specialized history skill.

  • Handoff

    ggoosen/Digital-Workforce/.claude/skills/handoff Skill

    1★ repo

    Compact the current conversation into a handoff document so a fresh agent (or a future you) can pick up the work without re-deriving anything. Writes to the OS temp directory — never the workspace — and references existing artifacts (the context wiki, outputs, inbox, raw) by path instead of duplicating them. Use when a session is getting long, before a context reset, or when you want to pause and resume later. Voice triggers — "handoff", "hand off", "pause and resume", "write a handoff", "compact this".

  • Geo audit

    ifitsmanu/landing-studio/skills/geo-audit Skill

    1★ repo

    Audit how a brand and its landing pages are retrieved, cited, linked, or misrepresented across generative search and answer products such as Google AI features, ChatGPT Search, Claude search, Perplexity, Gemini, and Copilot. Use for GEO, AI search visibility, LLM citations, crawler access, robots posture, entity consistency, answer accuracy, competitor share of citations, llms.txt, or agent-friendly site questions. Verify each vendor's current official docs, measure a dated prompt panel, keep SEO as the foundation, and return a sourced plan. Do not promise special AI ranking hacks, mandatory chunking, or llms.txt benefits that a target engine has not documented.

  • Use blender mcp

    wjlgatech/design-anything/skills/use-blender-mcp Skill

    no license1★ repo

    Drive blender-mcp for design-anything work — Drive Blender from any LLM via MCP — the natural execution backbone from agent to printable geometry. Use when a design task routes to this tool (agent-infrastructure). GENERATED satellite skill: facts fetched 2026-07-19; treat as a pointer, not gospel.

  • Nemo rl session memory

    junjunjunbong/research-loop/vendor/nvidia/nemo-rl-session-memory Skill

    no license1★ repo

    Manage durable working-session memory for coding agents. Use when a user asks to preserve or recover agent context across disconnects, VS Code restarts, long-running work, handoffs, or any session where important state should be written periodically under the repo's session directory. Do NOT use for: simple questions, short tasks, one-off commands, linting, or code review.

    The same file is in 1 other repository

  • Eval audit and sweep

    YosukeIida/personal-agent-skills/eval-audit-and-sweep Skill

    no license1★ repo

    This skill should be used when a user wants to (a) audit the quality and reliability of an existing LLM evaluation suite, or (b) determine which Claude model and inference parameters give the best quality-per-dollar and quality-per-second for their specific task by running a parameter sweep over that eval. Applicable to any eval framework (custom harnesses, tau-bench, inspect-ai, pytest-based, etc.) since the guidance is framework-agnostic.

  • Author

    baldaworks/hostplugin/plugins/hostplugin/skills/author Skill

    1★ repo

    Author, audit, or update portable plugins for Codex, Claude Code, Grok Build, GitHub Copilot CLI, OpenCode, and Cursor. Use when an agent must inspect a project or CLI, design host-native skills, commands, agents, hooks, MCP/LSP integrations, rules, or apps, scaffold either a standalone marketplace or an embedded plugin, and preview every file before writing.

  • Handoff

    yyqqCoding/engineering-flow-skills/skills/handoff Skill

    1★ repo

    Create a compact continuation record for another session or agent.

  • Claude md forge

    fullREFIT/skills/skills/claude-md-forge Skill

    1★ repo

    Turn source material — a Claude or ChatGPT conversation, a YouTube or podcast transcript, a strategy doc, notes — into a topic-specific agent context file (CLAUDE.md for Claude Code, AGENTS.md for Codex CLI and Cursor, or AGENTS.md plus a thin CLAUDE.md shim for a shared repo), so every future session in that directory operates with the strategy, tactics, vocabulary, and rejected paths from that source. Distills rather than summarizes: keeps only what changes agent behavior, pushes depth to docs/, refuses to produce README-style documentation. MANDATORY TRIGGERS: CLAUDE.md, AGENTS.md, agents md, claude md, turn this transcript into a CLAUDE.md, make an AGENTS.md from this conversation, convert this chat into project context, capture this session's knowledge, topic CLAUDE.md, domain context file, persistent context from transcript, YouTube transcript to CLAUDE.md, seed a project with this knowledge, agent memory from this conversation, claude-md-forge.

  • Claude release guide

    fullREFIT/skills/skills/claude-release-guide Skill

    1★ repo

    Convert shallow source material about a new Claude product or feature release (YouTube transcripts, blog posts, docs, announcements, pasted text) into a verified, teaching-grade implementation guide with prerequisites, exact steps, decision frameworks, edge cases, and cited sources. Specialized sibling to transcript-deep-dive: adds primary-source verification against claude.com and support docs, cross-references multi-source input for contradictions, outputs guides for teaching (Skool, LinkedIn, lead magnets, workshops) Use for Claude Design, Claude Code, Claude Cowork, Skills, MCP, Opus/Sonnet/Haiku model drops, pricing changes, platform features. MANDATORY TRIGGERS: claude release guide, new claude feature, claude release, claude announcement, claude product guide, teach a claude feature, claude how-to, new opus model, new sonnet model, new haiku model, claude design guide, claude code release, claude cowork release, release breakdown, feature deep dive.

  • Gpt 5 6 relay

    fullREFIT/skills/skills/gpt-5-6-relay Skill

    1★ repo

    Route one task through model-specific GPT-5.6 child threads with explicit thinking efforts and checkable handoffs.

  • Logo designer

    fullREFIT/skills/skills/logo-designer Skill

    1★ repo

    Professional logo and visual identity design system that translates brand strategy into optimized AI image generation prompts for external tools. Creates comprehensive visual identity concepts using methodologies from 11 elite design studios. Generates tool-specific prompts for Midjourney, DALL-E, Ideogram, Flux, Freepik, and Google Nano Banana Pro. Use when you need logo prompts to run in external AI tools rather than direct in-session generation. MANDATORY TRIGGERS: logo prompts, AI logo prompts, logo concepts, visual identity prompts, Midjourney logo, DALL-E logo, Ideogram logo, Flux logo, logo prompt suite, logo prompt engineering, logo ideas, prompt for logo, logo prompt.

  • Plan execute router

    fullREFIT/skills/skills/plan-execute-router Skill

    1★ repo

    Executes the plan/execute model routing pattern to cut AI-workflow cost by 50–75% on feature builds, refactors, and migrations. Claude plans the task at frontier capability and produces a detailed spec, then routes execution to a cheaper model — via Codex CLI subprocess if available, a Claude Code subagent scoped to a smaller model, or manual copy-paste instructions when neither path exists. Optionally invokes a frontier review pass on the executor's output. Includes fit-check to detect when a task does NOT fit the pattern and cost estimation for the specific split chosen. MANDATORY TRIGGERS: model routing, plan-execute, plan then execute, route to cheaper model, delegate execution, split planning from execution, cheaper executor, cost split, plan-execute-router, model split, use Codex for execution, use Sonnet for execution, spec-then-code, spec then code, save on this build, cost-optimize this task.

  • Project doc refresh

    fullREFIT/skills/skills/project-doc-refresh Skill

    1★ repo

    Audit, research, and refresh Claude Project custom instructions and project documents so they reflect current Claude platform capabilities (Skills 2.0, MCP, subagents, caching, model routing) and current domain facts. Use when the user wants to update a CLAUDE.md, custom instructions, project documents, system prompts, or any long-lived context document that may have gone stale. Three-phase workflow — audit first, research only flagged items, produce a tracked-changes diff the user approves before finalizing. Never rewrites voice, behavioral rules, or institutional knowledge. MANDATORY TRIGGERS: refresh custom instructions, update project documents, audit CLAUDE.md, update system prompt, refresh project context, is my prompt stale, update my Claude project, modernize instructions, refresh skill docs, project doc audit, instructions audit, context refresh.

  • Project orchestrator

    fullREFIT/skills/skills/project-orchestrator Skill

    1★ repo

    Multi-platform AI project orchestrator that decomposes projects into routed task plans across Claude Cowork, Claude Code, Cursor, Warp.dev, n8n, and other AI tools. Use when the user provides project details (website build, brand identity, HR campaign, product launch, etc.) and needs an execution plan distributed across multiple AI platforms. Triggers on phrases like "orchestrate this project", "plan this across tools", "break this down", "create a task plan", "route this project", or any multi-phase project that would benefit from cross-platform execution. Also triggers when user provides a project brief, PRD, or scope document and wants it turned into actionable task files.

  • Skill creator

    fullREFIT/skills/skills/skill-creator Skill

    1★ repo

    Guide for creating effective skills that extend agent capabilities with specialized knowledge, workflows, or tool integrations. Use this skill when the user asks to: (1) create a new skill, (2) make a skill, (3) build a skill, (4) set up a skill, (5) initialize a skill, (6) scaffold a skill, (7) update or modify an existing skill, (8) validate a skill, (9) learn about skill structure, (10) understand how skills work, or (11) get guidance on skill design patterns. Trigger on phrases like "create a skill", "new skill", "make a skill", "skill for X", "how do I create a skill", or "help me build a skill".

  • Skill translate

    fullREFIT/skills/skills/skill-translate Skill

    1★ repo

    Port a skill from one AI coding or chat platform to another. First checks whether translation is needed at all: Claude Code, Codex CLI, Cursor, and other Agent Skills-standard tools take the same SKILL.md unchanged, so those are installs, not conversions — copy the folder and validate. Genuine translation applies only to platforms with proprietary formats: ChatGPT Custom GPTs plus Actions, Gemini Gems and GEMINI.md, Perplexity Spaces, Hermes Agent. Maps each component to the target's character limits, MCP support, and file format. MANDATORY TRIGGERS: skill translate, convert skill, port skill, install this skill everywhere, install skill in codex, install skill in cursor, sync my skills across tools, does this skill work in cursor, skill for chatgpt, skill for gemini, skill for perplexity, skill for hermes, translate to chatgpt, translate to gemini, cross-platform skill, make this work in chatgpt.

  • Vibe coding router

    fullREFIT/skills/skills/vibe-coding-router Skill

    1★ repo

    Route vibe-coding work to the right agent skill/tool without loading a mega-prompt. Use when you are building, planning, debugging, QAing, shipping, or content-mining software with Claude Code, Codex, or another LLM; when a request mentions vibe coding, super-app, command center, skill routing, design-taste, superpowers, Agent-Native, Mastra, or choosing which installed skill/tool should drive the next phase. MANDATORY TRIGGERS: vibe coding, vibe-coding-router, super-app, command center, choose skills, route skills, phase router.

  • Entalpa implement

    entalpa/entalpa-mcp/skills/entalpa-implement Skill

    1★ repo

    Use this skill when working with Entalpa through its MCP server to implement stories and requirements that were authored on the Entalpa website, inspect the project/stakeholder/story/requirement/subsystem hierarchy, record code references, implementation status, and test coverage back onto requirements, or otherwise perform requirements-engineering work that should preserve traceability and avoid destructive MCP tool calls.

  • Entalpa prd

    entalpa/entalpa-mcp/skills/entalpa-prd Skill

    1★ repo

    Use this skill to generate a human-readable PRD/BRD document artefact from an Entalpa project over its MCP server — a stakeholder-facing spec (business-requirements narrative, per-subsystem functional and non-functional requirements with stable codes, data & storage and Definition-of-Done tables, and story/requirement traceability) exported as Markdown and, optionally, PDF. Covers the whole project or a selected subset of subsystems/stories. It reads project data and writes a document; it does not modify the project unless the user explicitly asks.

  • Thesis production orchestrator

    lihongqing2001-gif/chinese-thesis-production-skills/skills/thesis-production-orchestrator Skill

    1★ repo

    Guide and orchestrate an evidence-first Chinese STEM degree-thesis project from first-run intake through submission. Use when a new Codex user starts, resumes, restructures, plots, formats, revises, recovers, or audits a thesis; when the authoritative manuscript is unclear; or when literature, methods, data analysis, Origin plots, chapters, figures, citations, Word, reviews, and optional computation must be coordinated.

  • Design agent capability

    ImYourBoyRoy/agent-continuity-stack/skills/design-agent-capability Skill

    1★ repo

    Design or consolidate durable agent capabilities: instructions, skills, scripts, integrations, hooks, workflows, subagents, or bundles. Use for repeated workflows or overlap; inventory existing capabilities first.

  • Maintain agent instructions

    ImYourBoyRoy/agent-continuity-stack/skills/maintain-agent-instructions Skill

    1★ repo

    Create, audit, or edit required project agent instructions and durable Codex, Cursor, Claude Code, or Antigravity guidance. Use for missing AGENTS.md, global contracts, CLAUDE.md, and Rules while preserving precedence, privacy, and rollback.

  • Maintain project memory

    ImYourBoyRoy/agent-continuity-stack/skills/maintain-project-memory Skill

    1★ repo

    Create, audit, compact, or synchronize the required MEMORY.md and related continuity artifacts. Use after meaningful project work, before a resumable handoff, or when project memory is missing, stale, unsafe, or duplicative.

  • Eval contract writer

    authenticfake/clike/extensions/vscode/templates/harper-init/.clike/skills/eval-contract-writer Skill

    1★ repo

    Require every runnable KIT to emit deterministic validation artifacts for EVAL and GATE.

  • Skill template advanced

    JarrydAdaens/agentic_rails_tooling/context/templates/skills/skill-template-advanced Skill

    1★ repo

    Use this template when creating a broad Agent Skill that needs progressive disclosure, focused reference files, routing guidance, examples, and validation without loading all domain knowledge up front.

  • Skill template

    JarrydAdaens/agentic_rails_tooling/context/templates/skills/skill-template Skill

    1★ repo

    Use this template when creating a new Agent Skill that needs valid SKILL.md frontmatter, concise trigger guidance, workflow instructions, examples, and validation notes.

  • Rails memory learnings

    JarrydAdaens/agentic_rails_tooling/source/skills/rails-memory-learnings Skill

    1★ repo

    Use this skill at the end of a work session - after implementation, planning, Agentic Rails implementation-plan execution, bulk planning, orchestration, or a completed story - when the user asks to capture learnings, memorize the session, write a session learning, or deposit a memory artifact. It reflects on the session and writes exactly one durable learning file into the Rails memory repository's learnings/new/ folder, recording the outcome, what went well, what struggled, and full provenance: the models, harness, thinking effort, reasoning vs non-reasoning, fast mode, and token cost behind the work.

  • Rails new proposal

    JarrydAdaens/agentic_rails_tooling/source/skills/rails-new-proposal Skill

    1★ repo

    Use this skill when the user wants to capture a proven, project-local innovation - a guard rail, hook set, agent, rule pack, script, or workflow - as a self-contained proposal for the Agentic Rails tooling repository. Triggers: "make this a proposal", "propose this for rails", "bundle this up for tooling", "this guard rail should be generalized". It copies the working slice verbatim (hooks extracted into a fresh .claude/settings.json, agents, scripts, dependency manifests, docs) into proposals/<slug>/ in the tooling repo, writes a README recording the use case, the evidence it works, and a suggested destination, and can zip the proposal for sharing with another developer. It is drop-off only: it never integrates anything into source/, never generalizes code, and never touches other proposals.

  • Worktree

    HikaruEgashira/agent-skills/wf/skills/worktree Skill

    1★ repo

    Planが承認/完了した直後に自律的に呼び出す必要があるスキルです。 Trigger: plan approved, plan completed, taskを開始します

    The same file is in 3 other repositories

  • Claude code cli

    yu-iskw/coding-agent-skills/skills/claude-code-cli Skill

    no license1★ repo

    Executes tasks using the Claude Code CLI (`claude`). Automatically determines the safest permission mode (plan, acceptEdits, auto, or bypassPermissions) based on the task type.

  • Copilot cli

    yu-iskw/coding-agent-skills/skills/copilot-cli Skill

    no license1★ repo

    Executes tasks using the GitHub Copilot CLI (`copilot`). Automatically determines the safest permission mode (plan, editor, or autopilot) based on the task type.

  • Cursor agent cli

    yu-iskw/coding-agent-skills/skills/cursor-agent-cli Skill

    no license1★ repo

    Executes tasks using the Cursor Agent CLI (`cursor-agent`). Provides structured modes for planning, Q&A, and full automation with safety guardrails.

  • Llm debugger

    bob798/ai-skill-kit/ai-engineering/llm-debugger Skill

    no license1★ repo

    LLM 应用生产级问题排查 Skill。当用户提到"LLM 接口报错"、"响应太慢"、"token 超限"、"流式输出中断"、"API 调用失败"、"LLM 输出不稳定"、"上线后出问题"、"怎么监控 LLM 应用"时触发。适用于 LLM 应用开发调试、生产问题排查、系统稳定性优化。

  • Prompt optimizer

    bob798/ai-skill-kit/ai-engineering/prompt-optimizer Skill

    no license1★ repo

    系统化 Prompt 诊断与优化 Skill。当用户提到"这个 Prompt 效果不好"、"LLM 输出总是不对"、"模型不听指令"、"输出格式乱"、"回答太长/太短"、"怎么写好 Prompt"、"帮我优化这个 Prompt"时触发。适用于 System Prompt 设计、任务 Prompt 调试、Agent 指令优化等场景。

  • Rag evaluator

    bob798/ai-skill-kit/ai-engineering/rag-evaluator Skill

    no license1★ repo

    RAG pipeline 系统评估 Skill。当用户提到"评估我的 RAG"、"RAG 效果不好"、"知识库检索不准"、"大模型回答错了"、"RAG 怎么优化"、"chunk 怎么分"、"向量检索没召回"时触发。适用于 RAG 系统搭建、调试、优化全阶段。

  • Vector db guide

    bob798/ai-skill-kit/ai-engineering/vector-db-guide Skill

    no license1★ repo

    向量数据库选型、使用与调优 Skill。当用户提到"向量数据库怎么选"、"Chroma 和 Milvus 有什么区别"、"embedding 怎么存"、"相似度搜索怎么做"、"向量检索太慢"、"RAG 用哪个向量库"时触发。适用于 RAG 系统搭建、向量数据库选型、面试向量数据库考题。

  • Eval debug

    bob798/ai-skill-kit/development/eval-debug Skill

    no license1★ repo

    从开发视角分析面试 JSONL 文件,评估模型可用性、评分准确性、程序逻辑正确性,输出 bug 报告和改进建议。

  • Rag eval

    bob798/ai-skill-kit/development/rag-eval Skill

    no license1★ repo

    分析 RAG 实验结果 JSON 文件,解释指标含义,识别异常,给出一句话结论。支持 v4-v9 的所有结果文件。

  • Claude scheduler

    bob798/ai-skill-kit/workflow/claude-scheduler Skill

    no license1★ repo

    Schedule recurring Claude prompt tasks via the system scheduler (launchd on macOS; systemd --user or crontab on Linux). Invoke when the user asks for a recurring Claude task — triggers include "每天/每周/每小时", "定时", "schedule a claude task", "daily report", "weekly review", "run this prompt every N days", "后台跑 Claude". Claude creates, lists, and removes tasks autonomously through the `scheduler` CLI. Do not use for plain shell cron jobs — route those to cron/at directly.

  • Session state

    jovd83/playwright-skill/documentation/session-state Skill

    1★ repo

    Use when Playwright work spans multiple sessions or operators and a live resume pointer is needed. Maintain a current state file that records the latest completed step, current in-progress step, remaining work, blockers, touched files, resume commands, validation snapshot, and links to the latest handover.

  • Orchestrator

    jovd83/playwright-skill/orchestrator Skill

    1★ repo

    Central entrypoint for broad or ambiguous Playwright requests. Use when Codex needs to classify the user's testing goal, choose the right Playwright subskill, and move from intent to implementation, planning, documentation, execution, or reporting without unnecessary menu-driven back-and-forth.

  • Llm inference

    samarth777/modal-skills/skills/llm-inference Skill

    1★ repo

    This skill teaches Claude how to effectively build and deploy applications on Modal's serverless platform.

  • Openspec superpower orchestration

    fightZy/simple-skills/.agents/skills/openspec-superpower-orchestration Skill

    1★ repo

    Use when a non-trivial change should follow an OpenSpec-plus-superpower workflow with formal specs, approval gates, implementation discipline, and no parallel planning system

  • Mcp builder

    MatrixFounder/Universal-skills/skills/mcp-builder Skill

    1★ repo

    Guidelines for creating high-quality MCP (Model Context Protocol) servers. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).

  • Skill creator

    MatrixFounder/Universal-skills/skills/skill-creator Skill

    1★ repo

    Use when creating new Agent Skills, upgrading existing skills, running evals to test a skill, benchmarking skill performance, or optimizing a skill's description for better triggering accuracy. Guidelines for Gold Standard skill structures.

  • Agentprivacy dual agent harness

    mitchuski/agentprivacy-skills/agentprivacy-skills-v5/meta/agentprivacy-dual-agent-harness Skill

    1★ repo

    The operational Swordsman ⊥ Mage autoresearch loop: how to actually RUN dual-agent separation as a self-improvement harness against a concrete optimization target. The Mage (bnot · proposer) conceals/reduces; the Swordsman (neg · prover) signs only what survives an un-tuneable held-out gate; the Gap between them is the non-collusion precondition, physically realized as a Fiat-Shamir witness set the proposer cannot see. Activates when building or driving an agent loop that must produce VALIDATED improvements (circuit/gate/qubit reduction, cost optimization, any "make it cheaper and prove it" task) without fooling itself with mirages. Triggers: "dual-agent loop", "swordsman mage harness", "autoresearch", "proposer prover loop", "reduce gates or qubits", "ecdsa.fail", "circuit co-evolution", "held-apart agents", "neg bnot succ". V6 register note (2026-06-10): conjecture/version citations resolve to agentprivacy-docs/research/CONJECTURE_REGISTER_V6.md (head C89); model head: privacy_value_v6_formal_specification.md.

  • Agentprivacy agent interop

    mitchuski/agentprivacy-skills/agentprivacy-skills-v5/role/agentprivacy-agent-interop Skill

    1★ repo

    Agent interoperability protocols for 0xagentprivacy multi-agent systems. Activates when discussing M(u,y) matching functions, agent-to-agent communication, ERC-8004 trustless agent identity, cross-agent coordination, or how swordsman and mage agents interact across boundaries.

  • Agentprivacy ai agent

    mitchuski/agentprivacy-skills/agentprivacy-skills-v5/role/agentprivacy-ai-agent Skill

    1★ repo

    AI agent architecture and dual-agent separation for 0xagentprivacy. Activates when discussing I(S;M|π) mutual information bounds, TEE isolation, agent lifecycle (spawn/operate/rotate/recover/retire), Oracle architecture, or the fundamental design of privacy-preserving AI agents. V6 register note (2026-06-10): conjecture and version citations resolve to agentprivacy-docs/research/CONJECTURE_REGISTER_V6.md (head C89); model head: privacy_value_v6_formal_specification.md.

  • Agentprivacy wiki claude ghost

    mitchuski/agentprivacy-skills/agentprivacy-skills-v5/wikis/agentprivacy-wiki-claude-ghost Skill

    1★ repo

    A ghost page whose content is written live by Claude via the Anthropic API, inserted between a wiki form submission and the page-JSON response. Activates when building a Claude-authored ghost endpoint, a /claude-ghost FastAPI handler, or live LLM page generation in a fedwiki. Kept by the Librarian 🗃️ in the Tower's Wikis.