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
KhaledSaeed18/dotclaude/.claude-plugin/plugins/engineering/skills/parallel-agents Skill
4★ repoFan independent work out to multiple subagents that run concurrently, each with a focused scope and self-contained instructions, then review and integrate their results. Use when you face two or more genuinely independent tasks — separate failing test files, unrelated bugs, distinct subsystems — that share no state and don't depend on each other's order.
KhaledSaeed18/dotclaude/.claude-plugin/plugins/productivity/skills/handoff Skill
4★ repoCompact the current conversation into a handoff document for another agent to pick up. Use when ending a session, switching agents, or preserving context before compaction.
KhaledSaeed18/dotclaude/skills/engineering/create-agent Skill
4★ repoAuthor a new subagent for this repository end to end by scaffolding it with pnpm new, curating its tool allowlist, setting model, color, and memory in frontmatter, then writing a focused system prompt and regenerating the registry. Use when creating, scaffolding, or reviewing an agent or subagent in this repo.
KhaledSaeed18/dotclaude/skills/engineering/create-hook Skill
4★ repoAuthor a new Claude Code hook for this repository end to end by scaffolding it with pnpm new, writing the hook script and its settings.json wiring, documenting activation in HOOK.md, then regenerating the registry. Use when creating, scaffolding, or reviewing a hook in this repo.
KhaledSaeed18/dotclaude/skills/engineering/parallel-agents Skill
4★ repoFan independent work out to multiple subagents that run concurrently, each with a focused scope and self-contained instructions, then review and integrate their results. Use when you face two or more genuinely independent tasks — separate failing test files, unrelated bugs, distinct subsystems — that share no state and don't depend on each other's order.
KhaledSaeed18/dotclaude/skills/productivity/handoff Skill
4★ repoCompact the current conversation into a handoff document for another agent to pick up. Use when ending a session, switching agents, or preserving context before compaction.
mikestangdevs/craft-skills/skills/rigor/checkpoint-handoff Skill
4★ repoUse before any context boundary — a context compaction, the end of a long session, a handoff to another agent or a future session — on multi-session work. Writes the resume state DOWN, outside the conversation: a working doc with what's done, what's in flight, the numbered remaining tasks in order, the standing constraints (the rules that must survive verbatim), and the files worth re-reading first. The test: a cold session with zero conversational memory could pick up the next task without asking a single question. Use when the user says they're about to compact, when context is getting long mid-project, when pausing a phased effort, or when work will continue 'tomorrow' (it never continues in the same window).
izyanrajwani/agent-skills-library/skills/subagent-driven-development Skill
no license4★ repoSequential subagent execution with two-stage review gates for implementation plans. Use when executing multi-task plans in current session, when tasks need fresh subagent context to avoid pollution, when formal review cycles (spec compliance then code quality) are required between tasks, or when you need diff-based validation of each task before proceeding.
The same file is in 8 other repositories
dysfunc/ai-plugins-cc/dist/plugins/grok/skills/grok-prompting Skill
4★ repoInternal guidance for composing Grok prompts for planning, research, diagnosis, and review tasks inside the Grok Claude Code plugin
dysfunc/ai-plugins-cc/plugins/gemini/skills/gemini-prompting Skill
4★ repoInternal guidance for composing Gemini prompts for planning, research, diagnosis, and review tasks inside the Gemini Claude Code plugin
dysfunc/ai-plugins-cc/plugins/grok/skills/grok-prompting Skill
4★ repoInternal guidance for composing Grok prompts for planning, research, diagnosis, and review tasks inside the Grok Claude Code plugin
akhilkannur/marketing-agent-blueprints/skills/agent-context-builder Skill
no license4★ repoAI agents struggle when they don't know your project's structure. This agent scans your current directory, identifies key files/stacks, and generates an `AGENTS.md` file. This file acts as a 'ReadMe for Robots,' ensuring every future agent knows exactly how to work with your code.
tuoxie2046/claude-code-research-skills/skills/autofigure Skill
no license4★ repoGenerate clean, EDITABLE vector figures (SVG + exact-size PDF) for research papers — method / architecture / pipeline / system-overview diagrams — with AutoFigure-Edit. Use whenever the user wants to create or vectorize a paper figure from a text description OR from a draft/screenshot/draw.io image: it generates a step-1 raster with OpenAI gpt-image-2 by default (nano-banana optional), then runs local SAM3 segmentation + RMBG-2.0 icon extraction + an LLM SVG re-draw with OpenAI gpt-5.5 into an editable SVG, and exports a margin-free PDF for \includegraphics. Triggers: make a paper figure, method/architecture/pipeline/framework diagram, vectorize a diagram, "editable SVG figure", "用 AutoFigure / AutoFigure-Edit 画图", 论文方法图/流程图/架构图.
tuoxie2046/claude-code-research-skills/skills/gpt-image Skill
no license4★ repoGenerate, edit, describe, or batch-generate images from the terminal using OpenClaw + OpenAI gpt-image-2 (billed to the ChatGPT/Codex subscription, not the API). Use whenever the user asks to create / draw / generate an image, picture, icon, logo, illustration, poster, diagram, architecture figure, flowchart, or infographic — to produce a set / batch of multiple images from several prompts — or to edit an existing image (inpaint / restyle / combine) or describe what is in an image file.
davccavalcante/supreme-coding-guidelines-skill.ah/skills/supreme-benchmarking Skill
no license4★ repoPrincipal research and data-science benchmarking discipline for AI, ML, LLM, and npm/Node projects, inspired by the published methodologies of OpenAI (simple-evals, SWE-bench Verified audits), Anthropic (model cards with methodology appendix and error bars), Google DeepMind (benchmark tables with disclosure appendix), xAI (live benchmarks with explicit cutoff dates), DeepSeek (radical transparency — full hyperparameters, compute, distillation recipes), Xiaomi MiMo (pass-at-1 averaged over many seeds), Hugging Face (Open LLM Leaderboard normalization, lighteval, versioned harnesses), and Unsloth (efficiency benchmarks with reproducible notebooks). Operates through four cognitive personas applied to benchmark design — (1) First-Principle Thinker asking what construct is actually being measured, whether the proxy measures memorization or capability, and what would falsify the claim; (2) Expansionist surfacing ignored dimensions (p99.9 latency, cold start, cost per task, energy, robustness to paraphrase, multi-turn degradation, variance across seeds); (3) Outsider removing bias via preregistration of methodology before results, blind evaluation where possible, baseline re-run integrity (never copy competitor numbers from papers — re-run all baselines in the same config), and the test of whether a competitor or regulator would accept the methodology; (4) Executor running the controlled protocol peer-to-peer without trying to please — pinned versions, fixed seeds, minimum five runs, warmup discarded, isolated environment, raw data published, and uncomfortable results published unchanged. Enforces statistical rigor gates (95 percent confidence intervals, median plus IQR, effect size, significance tests, multiple-comparison correction), contamination and saturation defense (canary strings, n-gram overlap checks, held-out private sets, benchmark refresh cadence), LLM-specific protocol (pass-at-k disclosure, temperature disclosure, prompt-sensitivity spread across three to five paraphrases, eval-harness version pinning — lm-eval-harness, HELM, lighteval, inspect-ai, promptfoo), npm/Node-specific protocol (tinybench or mitata for micro, hyperfine for CLI, size-limit for bundle, JIT warmup and GC isolation), reporting in the Anthropic, Hugging Face, and Unsloth style (tables with best result bolded, bar charts with error bars, radar charts for multi-dimension, efficiency-frontier scatter of score versus cost, methodology appendix, benchmark card analogous to a model card, raw JSONL published), one-command reproducibility package, honest disclosure (negative results published, best-of-N disclosed, limitations section, conflicts of interest declared), and continuous regression tracking in CI with threshold alerting. Requires ah-parser. Output mode follows user preference at parser activation; user code, eval outputs, raw data, configs, and benchmark artifacts are always preserved verbatim. Honest measurement over impressive numbers — non-negotiable.
JWCodeWrote/Agent_Skills_Plugin/agent-fault-retrospective Skill
no license4★ repoUse when a user identifies an AI agent fault, overreach, misunderstanding, unauthorized action, or asks for a retrospective, durable fault rule, trigger keyword, or future automatic invocation; also use when existing fault records need to be consulted.
JWCodeWrote/Agent_Skills_Plugin/frontend-imagegen-director Skill
no license4★ repoUse when the user wants the agent to inspect the current project's style and product nature, clarify requirements in a Superpowers-style workflow, generate front-end visuals with gpt-image2 or the environment's imagegen path, and then build the actual front-end page from the approved image.
indykite/skills/indykite-mcp-server Skill
4★ repoMake live IndyKite authorization decisions (AuthZEN/KBAC) and run ContX IQ graph queries from an AI agent over the Model Context Protocol - one Bearer-token JSON-RPC session, no bespoke REST wiring. Use when initializing an MCP session, calling its tools (authzen_evaluate, authzen_evaluations, authzen_search_*, ciq_execute), configuring an MCP server, or debugging its single Bearer-token auth.
cfregly/gpu-perf-tune/plugins/profile-and-optimize/skills/inference-model-eval Skill
4★ repoDrive lm-eval-harness quality evals (GPQA, MMLU-Pro) inside model pods plus optional ExternalEval (externally operated). Use to validate model quality before promoting to staging or prod, after vLLM / quantization / KV-cache changes, or to compare against published HuggingFace / paper baselines. Pair with inference-perf-bench (the perf-side counterpart) for full pre-promotion validation. Triggers on "lm-eval-harness", "GPQA", "MMLU-Pro", "ExternalEval", "model quality eval", "/run-model-eval", "run evals", "run gpqa", "run mmlu", "run external-eval", "run evals on the model", or any combination of "eval / quality / accuracy" with "inference / model / vllm".
nataliacorrea03/claude-code-skills/skills/meta-prompt Skill
4★ repoBuilds ready-to-paste Meta AI (Muse Spark) prompts for reading Instagram, Facebook, and Threads directly. Describe a research goal (find micro-influencers or ambassadors, find videos to react to, monitor brand mentions, scan niche trends, check saturation, mine comments, tear down a competitor, verify handles, study a creator's voice before a pitch) and this skill asks a few clarifying questions, fills the variables (niche, follower range, brand, time window, etc.), and outputs a copy-paste prompt for the meta.ai app. It also hands out the Hook Machine, a full multi-turn workflow that analyzes a creator's top reels, extracts why the winners win, builds a grading rubric, then generates, grades, and rewrites hooks. It never calls Meta AI itself and never invents capabilities. Trigger on "/meta-prompt", "build a meta ai prompt", "meta prompt for <goal>", "find microinfluencers on instagram", "find videos to react to", "monitor mentions of <brand>", "what's trending on instagram for <niche>", "hook machine", "analyze a creator's hooks", "build a hook rubric", or any request to research Instagram creators, content, or brands that would be run inside Meta AI.
nataliacorrea03/claude-code-skills/skills/viral-hook-creator Skill
4★ repoGenerates OR grades viral social media hooks using a library of proven patterns, structures, a swipe file, and trigger words. Trigger on "/viral-hook-creator", "write viral hooks for <topic>", "draft hook options for <topic>" (GENERATE mode), or "grade this hook", "is this hook strong", "strengthen this hook", "score my hook", "make this hook better" (GRADE mode).
nataliacorrea03/claude-code-skills/skills/working-with-claude Skill
4★ repoAnthropic's official best practices for collaborating with Claude well, distilled from the Anthropic Academy courses. Covers the 4D framework (Delegation, Description, Discernment, Diligence), Claude's real capabilities and limitations, the prompting techniques that actually move quality, and the operating discipline for Claude Code. Use when starting a build / code / design / strategy task, when deciding what to hand to AI vs keep human, when a prompt is not landing, when evaluating whether Claude's output is trustworthy, or whenever you want to work with Claude the way Anthropic teaches. Complements build-mode (teach-while-building), frontend-design (UI aesthetics), and claude-api (API/model reference); this skill is the high-level "how to collaborate with Claude" layer.
keli-wen/skills/skills-zh/multi-agent-driven Skill
no license4★ repo把当前任务派发给多个 subagent——但先检查这次拆分到底值不值。触发语:「multi-agent this」「fan out」「用多智能体做」「派发出去」,或任务能拆成真正独立的子任务时。按上下文耦合度拆解,spawn prompt 用 handoff 的纪律写成任务契约。
keli-wen/skills/skills/handoff Skill
no license4★ repoGenerate a minimal-sufficient handoff when another agent or session will take ownership of a task or continue unfinished work. Use for /handoff, "把任务交给下一个 agent", "生成 handoff", "让另一个 session 继续", session transfer, task ownership transfer, or restarting in a fresh context when the current session has degraded (a handoff to your future self). Reuses durable artifacts by reference instead of copying them, and redacts secrets.
keli-wen/skills/skills/multi-agent-driven Skill
no license4★ repoDispatch the current task to multiple subagents — after checking that fan-out is actually justified. Use when the user says "multi-agent this", "fan out", "用多智能体做", "派发出去", or when a task decomposes into genuinely independent subtasks. Decomposes along context coupling, uses handoff discipline for spawn contracts.
megandmartin/agent-skills-repo/skills/agent-mastery/memory-hygiene Skill
4★ repoMaintains an agent's memory and notes files — dedupe entries, prune stale facts, convert relative dates to absolute, and verify claims are still true before re-asserting them, on a weekly sweep. Use when the user says "clean up memory", "the agent keeps repeating outdated info", "memory sweep", "my notes file is a mess", or on the scheduled weekly maintenance pass. Don't use for managing what loads into a live session's context — that's context-budget-manager.
megandmartin/agent-skills-repo/skills/agent-mastery/prompt-debugger Skill
4★ repoSystematic diagnosis when agent output misses the mark — isolate whether the fault is instructions, context, tools, or model; change one variable at a time; prove the fix with a before/after eval. Use when the user says "the agent keeps getting this wrong", "why is the output bad", "it ignored my instructions", "debug this prompt", or an eval case fails repeatedly. Don't use for building the eval that detects the failure — that's agent-eval-harness.
megandmartin/agent-skills-repo/skills/paperclip-workforce/delegation-protocol Skill
4★ repoCross-team delegation rules for Paperclip agents — when to delegate vs. do it yourself, how to write a delegable ticket (context, definition of done, deadline, budget), acceptance criteria for delivered work, and the retry → reassign → escalate-to-human ladder. Use when creating or assigning a ticket, deciding "should I delegate this", reviewing a delivered ticket, or handling a ticket that failed or stalled. Don't use for cEO-level portfolio planning (use ceo-operating-doctrine) or budget math (use budget-guardrails — this skill only requires that you did it).
megandmartin/agent-skills-repo/skills/paperclip-workforce/ops-agent-persona Skill
4★ repoPersona and protocol for the operations agent in a Paperclip org — scheduling, report compilation, file hygiene, and recurring-job monitoring; the org's reliability layer that surfaces every anomaly and never silently fixes one. Use when acting as the ops agent, when a ticket asks to "schedule", "compile the report", "clean up the files", "check the crons", or "why didn't the job run". Don't use for synthesizing the weekly retro narrative — that's weekly-retro.
megandmartin/agent-skills-repo/skills/paperclip-workforce/sales-agent-persona Skill
4★ repoPersona and protocol for the sales agent in a Paperclip org — works pipeline tickets, writes personalized outreach and follow-up DRAFTS (a human sends everything), keeps the pipeline-tracker CSV current, and never states or promises pricing without CEO approval. Use when acting as the sales agent, when a ticket asks to "work the pipeline", "draft outreach", "follow up with [prospect]", or "update the tracker". Don't use for existing-customer issues or complaints — that's support-agent-persona.
megandmartin/agent-skills-repo/skills/paperclip-workforce/support-agent-persona Skill
4★ repoPersona and protocol for the support agent in a Paperclip org — severity triage, calm plain-language tone rules, an always-escalate line for refunds, legal threats, and angry customers, and FAQ capture back into the knowledge base. Use when acting as the support agent, when a ticket routes a customer question, bug report, or complaint. Don't use for outbound prospect conversations — that's sales-agent-persona.
jerseycheese/agent-skills/skills/ci-fix-with-memory Skill
4★ repoAutomatically invoke when CI checks fail on a PR. Triggers on: any mention of CI failure, red checks, failed GitHub Actions, or "CI is broken/failing". Reads session handoffs and a known-issues log to avoid repeating failed fix attempts. Do NOT wait for user to ask — invoke as soon as a CI failure is confirmed. Not for local-only test failures (use test-fix instead) or lint-only failures with an obvious single-line fix.
jerseycheese/agent-skills/skills/project-bootstrap Skill
4★ repoBootstrap a project under ~/Projects with the standard agent scaffolding, stamped from the starter kit at shared/Development/templates/starter-kit - a CLAUDE.md from the house template, the stack-agnostic core (.editorconfig, .gitignore, .claude settings/launch, worktree-port script), an optional stack layer, and a row in the ~/Projects/CLAUDE.md root map. Use when creating a new project or repo, or when an existing repo is missing its CLAUDE.md. Trigger on "bootstrap this project", "new project setup", "stamp a CLAUDE.md", "set up agent scaffolding", "give this repo a CLAUDE.md".
jerseycheese/agent-skills/skills/worktree-enhanced Skill
4★ repoSets up a git worktree for isolated feature work — creates the branch, configures the worktree, detects whether a dev server is already running, and handles project-specific setup. Trigger on: "create a worktree", "set up a new branch", "start work on issue #X", "work on this in isolation", "new worktree for", "start a feature branch", "create an isolated environment for", "set up worktree".
Firzus/agent-skills/skills/game/figma-to-unity Skill
4★ repoImplements Figma designs (frames, components, HUD elements, menus) as Unity UI Toolkit interfaces — UXML hierarchy, USS styles mapped to design tokens, exported sprite assets, and a minimal C# controller — using the Figma MCP server (get_design_context, get_metadata, get_screenshot, get_variable_defs). Use when the user provides a Figma URL and wants it implemented in a Unity project, asks to "implement this Figma design in Unity", "build this HUD/menu/component from Figma", "convert Figma to UXML/USS", or mentions figma-to-unity.
Firzus/agent-skills/skills/web/astryx Skill
4★ repoAstryx (@astryxdesign) — Meta's open-source, agent-ready React + StyleX design system. Covers setup (npm packages, npx astryx init, StyleX compiler), the agent discovery loop (astryx template/component/docs, MCP server, --dense), per-component subpath imports, the xstyle styling order, design tokens, the Theme provider and defineTheme theming, cascade-layer CSS wiring, Tailwind and library interop, layout primitives, and incremental migration. Use when the user mentions Astryx, @astryxdesign, npx astryx, xstyle, defineTheme, or wants to set up, build UI with, theme, review, or migrate to the Astryx design system.
EnkrateiaLucca/skills/skills/resume-handoff Skill
4★ repoResume work from a handoff document written by a previous agent session, with context analysis and drift validation before any work starts. Use when the user says "resume from the handoff", "pick up where we left off", "continue from yesterday's session", "resume handoff", or points at a file in a handoffs/ directory. Pairs with the create-handoff skill, which writes the documents this skill consumes.
mhrsdev/AI-Agent-Skills-Library/organized-skills/code-quality-testing/git-guardrails-claude-code Skill
4★ repoSet up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute. Use when user wants to prevent destructive git operations, add git safety hooks, or block git push/reset in Claude Code.
The same file is in 18 other repositories
mhrsdev/AI-Agent-Skills-Library/organized-skills/code-quality-testing/mattpocock-skills/skills/misc/git-guardrails-claude-code Skill
4★ repoSet up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute. Use when user wants to prevent destructive git operations, add git safety hooks, or block git push/reset in Claude Code.
The same file is in 18 other repositories
mhrsdev/AI-Agent-Skills-Library/organized-skills/engineering-workflows/mattpocock-skills/skills/productivity/handoff Skill
4★ repoCompact the current conversation into a handoff document for another agent to pick up.
The same file is in 7 other repositories
mhrsdev/AI-Agent-Skills-Library/organized-skills/knowledge-learning-writing/.system/skill-creator Skill
4★ repoGuide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Codex's capabilities with specialized knowledge, workflows, or tool integrations.
The same file is in 3 other repositories
mhrsdev/AI-Agent-Skills-Library/organized-skills/product-planning-issues/handoff Skill
4★ repoCompact the current conversation into a handoff document for another agent to pick up.
The same file is in 7 other repositories
LichAmnesia/awesome-antigravity-skills/skills/skill-template Skill
4★ repoCopy-paste scaffold for creating a new Antigravity Agent Skill. Use when you want to author a skill folder that follows the official SKILL.md format.
Paldom/python-skills/skills/agent-guardrails Skill
4★ repoInstalls agentic guardrails into a Python repo — Claude Code hooks, a Stop gate that signs off only what pre-commit accepts, vetted public skills — and troubleshoots hooks that don't fire or block. Use for 'add Claude Code hooks', 'write an AGENTS.md', 'make this repo agent-ready', 'agent keeps bypassing checks', 'agent says done but pre-commit fails'. Not for authoring the pre-commit config.
Paldom/python-skills/skills/python-precommit Skill
4★ repoConfigures pre-commit for a Python repo — a .pre-commit-config.yaml with ruff, mypy, hygiene, commit-message and pre-push hooks, plus non-Python file formatting and validation (YAML, JSON, TOML, Markdown), kept in sync with CI. Use for 'set up pre-commit', 'add git hooks', 'format the YAML and Markdown', 'enforce conventional commits'. Not for Claude Code agent hooks or CI workflows.
agent-rig/rig/skills/rig-worktree Skill
4★ repoManage isolated git worktrees: create one wired up for dev (fetch, branch from base, symlink env, install deps), list them with PR state, or safely remove merged ones. A shared bootstrap that implement-style flows can call, also usable directly. Triggers on: 'worktree', 'set up a worktree', 'new worktree', 'list worktrees', 'remove worktree', 'isolate this in a worktree', 'symlink env into worktree'.
adsagents/adsagent-ai-skills/skills/agent-scheduled-tasks Skill
4★ repoAdsAgent tri-channel AI skills for Meta, Google Ads, and TikTok hosted MCP
oleg-koval/agent-skills/packages/software-development/promptctl Skill
4★ repoUse promptctl for template-based prompt engineering, prompt scoring, prompt generation, and reusable prompt workflows across repositories and teams.
oleg-koval/agent-skills/packages/software-development/promptctl/adapters/claude/skills/promptctl Skill
4★ repoUse promptctl for template-based prompt engineering, prompt scoring, prompt generation, and reusable prompt workflows across repositories and teams.
oleg-koval/agent-skills/packages/software-development/promptctl/adapters/cursor/skills/promptctl Skill
4★ repoUse promptctl for template-based prompt engineering, prompt scoring, prompt generation, and reusable prompt workflows across repositories and teams.
oleg-koval/agent-skills/packages/software-development/promptctl/adapters/grok/skills/promptctl Skill
4★ repoUse promptctl for template-based prompt engineering, prompt scoring, prompt generation, and reusable prompt workflows across repositories and teams.
Chris1Wang3/HammerRoom-Skills/skill-quality-scorer Skill
4★ repoDeterministic TRACE+ quality scorer for Agent Skills (SKILL.md): six dimensions T-R-F-S-I-E, 30 sub-items, merges TRACE, good-skill authoring reverse-rubric (Extra08/skill-creator), and ClawHub meta-skills (skill-reviewer, skill-quality-audit). Use when scoring, rating, auditing, or comparing skills; or says「给技能打分」「技能评分」「和同类 skill 差在哪」「是否符合规范」. Runs static_audit.py first; outputs JSON + Markdown. Not TRACE-only.
dceoy/ai-coding-agent-skills/skills/claude-command-converter Skill
4★ repoConvert Claude Code commands (.claude/commands/*.md) to standard Agent Skills (skills/*/SKILL.md). Use when migrating slash commands to portable skill format, creating skills from existing commands, or standardizing command definitions across AI runtimes.
dceoy/ai-coding-agent-skills/skills/code-review Skill
4★ repoPerform comprehensive code review on a pull request, checking for bugs, AGENTS.md/CLAUDE.md compliance, and historical context issues. Use when the user asks to review a PR or current changes.
vaquarkhan/compliance-agent-skills/skills/ccpa-cpra-privacy-rights Skill
4★ repoImplements California Consumer Privacy Act (CCPA) as amended by the California Privacy Rights Act (CPRA)—Cal. Civ. Code §1798.100 et seq.—covering consumer rights to know, delete, correct, opt-out of sale/share, and limit use of sensitive personal information, plus DSAR workflows, privacy notices, and service provider contracts for AI/agent data handling. Trigger when processing California residents' personal information, operating DSAR programs, reviewing agent/LLM data flows for CPRA compliance, or auditing privacy notices and vendor agreements. Do not use for HIPAA-covered PHI workflows (use hipaa-privacy-minimum-necessary), PCI cardholder encryption (use pci-dss-encryption-key-management), or general vendor SOC reports without CPRA scope (use vendor-third-party-risk).
vaquarkhan/compliance-agent-skills/skills/gdpr-us-multinational Skill
4★ repoImplements GDPR compliance workflows for US-headquartered multinationals—EU/EEA/UK data subjects, Articles 5-7 lawful basis, Article 30 records of processing, Article 32 security, Articles 33-34 breach notification (72 hours), Article 35 DPIA, Standard Contractual Clauses, EU-US Data Privacy Framework adequacy, and DPA/subprocessor governance for agent/LLM cross-border transfers. Trigger when US companies process EU residents' data, deploying agents in EU regions, assessing LLM vendor international transfers, or preparing RoPA/DPIA for multinational privacy programs. Do not use for US state laws only (use us-state-privacy-laws or ccpa-cpra-privacy-rights), HIPAA PHI (use hipaa-technical-safeguards), or UK-only post-Brexit without EU scope (note UK GDPR parallels in references).
vaquarkhan/compliance-agent-skills/skills/hipaa-baa-vendor-assessment Skill
4★ repoReviews Business Associate Agreements and subprocessors for LLM vendors, cloud providers, and MCP server operators under HIPAA (45 CFR §164.502(e), §164.504(e)). Trigger when onboarding OpenAI/Anthropic/Azure OpenAI, cloud hosts, observability tools, or MCP integrations that may access ePHI. Do not use for technical encryption testing (use hipaa-technical-safeguards) or general vendor SOC reports without BAA focus (use vendor-third-party-risk).
vaquarkhan/compliance-agent-skills/skills/hipaa-phi-redaction-pipeline Skill
4★ repoConfigures and validates the Presidio-based PHI redaction pipeline—DLP entity detection, reversible tokenization before LLM ingestion, and authorized deanonymization—integrated with redaction.py and the compliance agent. Trigger when ePHI may appear in prompts, implementing minimum-necessary LLM access, tuning entity types, or auditing redaction effectiveness. Do not use for HIPAA access control design (use hipaa-technical-safeguards) or legal BAA review (use hipaa-baa-vendor-assessment).
Hipaa privacy minimum necessary
vaquarkhan/compliance-agent-skills/skills/hipaa-privacy-minimum-necessary Skill
4★ repoImplements HIPAA Privacy Rule minimum necessary standard—45 CFR §164.502(b)—plus §164.514 de-identification, Limited Data Sets, and patient rights to access and amend records, with agent prompt minimization, role-based PHI exposure, and integration with the Presidio redaction pipeline. Trigger when designing LLM/agent PHI access policies, auditing disclosure practices, implementing patient rights workflows, or validating that agent prompts contain only minimum necessary ePHI. Do not use for Presidio entity tuning alone (use hipaa-phi-redaction-pipeline), Security Rule technical controls (use hipaa-technical-safeguards), or BAA legal review (use hipaa-baa-vendor-assessment).
vaquarkhan/compliance-agent-skills/skills/hipaa-technical-safeguards Skill
4★ repoImplements HIPAA Security Rule technical safeguards (45 CFR §164.312)—access control, audit controls, integrity, person/entity authentication, and transmission security—for AI agents, MCP servers, and LLM pipelines. Trigger when assessing ePHI handling in agent architectures, MCP tool authorization, encryption in transit/at rest, or mapping HIPAA controls to technical implementations. Do not use for BAA legal review (use hipaa-baa-vendor-assessment) or Presidio tokenization setup (use hipaa-phi-redaction-pipeline).