{"data":[{"slug":"rkstack","title":"rkstack","description":"Complete AI development workflow — 22 skills that guide the full cycle from idea to shipped code. Brainstorm a design, write an implementation plan, execute with TDD, debug systematically, get a two-pass code review, and ship with confidence. Includes a humanizer for natural prose, a security auditor (OWASP + STRIDE), and safety hooks that block destructive commands. Detects your project stack automatically and adapts.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/mrkhachaturov/rkstack.git"},{"slug":"riff","title":"riff","description":"Riff is a plugin made up of nine opinionated skills for writing personal essays with Claude. The idea behind it is to treat writing with AI as a process to be supported instead of a product to be created, and to keep the writer in charge at every step of the way. Together, the skills form a pipeline so a user can start working with raw notes and get to a finished piece, but each skill also encourages time for pausing and reflection when appropriate. A persistent style profile built from a user's own writing keeps drafts in their voice across sessions, and the plugin is designed to handle personal material with care. The goal throughout is to make writing with Claude more fun and effective.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/benjaroy/riff.git"},{"slug":"rex","title":"rex","description":"Rex provides sub-second Python symbol search across all installed .venv packages. Three tools — rex_find, rex_show, rex_members — let Claude look up function signatures, docstrings, class members, and file locations instantly, without web search or grepping through source files. Results are grounded in the actually installed version — no outdated docs, no hallucinated parameters. Auto-indexes on first use and re-indexes incrementally when packages change. Read-only by design.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/vykhovanets/rex.git"},{"slug":"revit-bim","title":"revit-bim","description":"Connect Claude to Autodesk Revit and Dynamo. Generate ready-to-run Dynamo Python scripts, query model data, detect clashes, audit model health, manage parameters, create schedules, export IFC, and save .dyn files — all from plain language. Includes 12 skills covering the full BIM workflow.\n\nCreated by Manas Bhatia (Architect and Design Technologist)","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/Utopia5327/revit-bim.git"},{"slug":"respira-for-wordpress","title":"respira-for-wordpress","description":"Edit live WordPress sites from inside Claude Cowork. Tell\nClaude what you want changed and Respira routes the edit to\nthe right page builder, snapshots before writing, and validates\nthat the page actually rendered correctly. Twelve page builders\nsupported (Elementor, Divi 4, Divi 5, Bricks, Oxygen, Breakdance,\nBeaver Builder, WPBakery, Flatsome, Brizy, Thrive Architect,\nGutenberg). One-click rollback always available. No FTP. No\ndatabase access. No REST plumbing.\n\nBuilt for the agency owners, designers, and freelancers who\nmanage WordPress sites for clients and want to delegate the\nediting work to Claude without giving up control.\n\nBundled: eight slash commands for the most common workflows,\nthirty auto-activating skills (including sixteen builder-to-\nbuilder migration paths), and one visual reviewer sub-agent\nthat opens the live page after every edit so you can see what\nchanged.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/respira-press/cowork-plugin-wordpress.git"},{"slug":"research-co-pilot","title":"research-co-pilot","description":"About\nComprehensive research co-pilot for Claude Code and claude.ai — 13 skills covering literature review, methodology, ethics review, data analysis, qualitative coding, brainstorming, manuscript drafting, replication design, grant writing, talk building, citations, survey design, and peer review. A peer collaborator, not a subservient assistant.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/Marazii/research-co-pilot.git"},{"slug":"repo-docs-system","title":"repo-docs-system","description":"Build and maintain a structured Chinese knowledge base (docs/ system) for any code repository. Generates AGENTS.md, ARCHITECTURE.md, design docs, product specs, and quality score reports.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/baiyue11111/repo-docs-system.git"},{"slug":"reflect","title":"reflect","description":"Session-local metacognition harness for long-running Claude Code work.\n\nWhen suggestions get reverted 3+ times within 10 tool calls, reflect invokes Opus 4.7 to reason about why the rejections clustered. The model reads back the last 20 tool calls, the rolled-back diff, and active rules, then writes structured guidance — pattern, signal, adjustment, confidence — to .reflect/session-guidance.md. The next turn auto-loads it via path-scoped rule injection and adjusts.\n\nKey features:\n• 3-tier weighted signal taxonomy: git revert/restore (Tier 1, +100), user-file deletes (Tier 2, +70), utterance negation regex (Tier 3, +50). Cumulative threshold 240 + 5-turn cooldown — fires only when reverts genuinely cluster.\n• Single-shot Opus 4.7 with adaptive thinking (effort: high) and 3-layer prompt cache (L1 1h, L2 5m, L3 ephemeral).\nMeasured: cold ~$0.05, warm ~$0.01, 95.9% L1 cache hit.\n• Two cross-platform hooks: PostToolUse (Tier 1+2) and UserPromptSubmit (Tier 3). Pure-bash arithmetic — no jq/bc dependency on Windows git-bash.\n• Session-local by design: no persistence, no cross-session sync, no telemetry. The guidance file evaporates at session end.\n• Failure modes characterized: cold-start refusal (< 5 turns), false-trigger flagging on intent shifts, abstention on regulatory/opaque domains.\n• Optional localhost Next.js companion Viewer (binds 127.0.0.1, never deploys, PII redactor on every read) for live state + reflection history + cross-session drift clustering.\n• MIT-licensed. SLSA v1 provenance via GitHub Actions OIDC trusted publisher (Sigstore-signed).","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/chanjoongx/reflect.git"},{"slug":"redis","title":"redis","description":"Official Redis plugin for Claude Code that helps AI coding agents build and optimize applications with Redis. It provides Redis development best practices across data modeling, key design, caching, query engine usage, vector search, semantic caching, connection handling, performance, security, and observability, so agents can generate better code and make safer architectural choices when working with Redis.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/redis/agent-skills.git"},{"slug":"real-time-backend","title":"real-time-backend","description":"Architects and implements production-grade backends with strong opinions on correctness and scalability. Designs schemas first, enforces type safety end-to-end, writes ACID transactions by default, and composes data server-side to eliminate request waterfalls. Real-time queries replace polling. Server functions replace REST boilerplate. Validated function signatures replace hand-written API contracts. Actively avoids common anti-patterns — N+1 queries, manual cache invalidation, layered archi...","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/get-convex/real-time-backend.git"},{"slug":"rayden-ui-design-skill","title":"rayden-ui-design-skill","description":"Generate React code and build Figma designs with the Rayden UI component library. Includes full component API reference, design tokens, craft rules, anti-patterns, and layout patterns","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/playbookTV/rayden-ui-design-skill.git"},{"slug":"qodo--community","title":"Qodo","description":"Qodo Skills provides a curated library of reusable AI agent capabilities that extend Claude's functionality for software development workflows. Each skill is designed to integrate seamlessly into your development process, enabling tasks like code quality checks, automated testing, security scanning, and compliance validation. Skills operate across your entire SDLC—from IDE to CI/CD—ensuring consistent standards and catching issues early.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/qodo-ai/qodo-skills.git"},{"slug":"qa-orchestra","title":"qa-orchestra","description":"10 specialized QA agents for Claude Code. Each agent answers a specific question about your PR: Does this diff implement the acceptance criteria? What test scenarios do I need? Which existing tests will break? and writes a structured Markdown report to qa-output/ that you can paste into GitHub, Jira, or Linear. Stack-agnostic: works with any web application (React, Angular, Vue, Rails, Django, Spring Boot, etc.). Agents include: functional-reviewer (diff vs AC analysis), test-scenario-designer, smart-test-selector, bug-reporter, browser-validator (live validation via Chrome MCP), environment-manager, automation-writer (Playwright/Cypress/Selenium), release-analyzer (multi-repo diffs), manual-validator, and an orchestrator that chains them together.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/Anasss/qa-orchestra.git"},{"slug":"python-clean-architecture","title":"python-clean-architecture","description":"Scaffold, review, and refactor Python projects with clean code and architecture. Design principles and patterns implemented the Pythonic way — using Protocol, Callable, functools.partial, and closures instead of heavy class hierarchies. Includes a /review-architecture command, layered FastAPI project scaffolding, and reference guides for 7 design principles, 17 code quality rules, and 10 Pythonic pattern recipes. Inspired by Arjan Codes.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/MKToronto/python-clean-architecture.git"},{"slug":"pwp","title":"pwp","description":"Project Workflow Protocol — 11 systematic skills for debugging, code review, security audit, refactoring, migration, API design, performance optimization, project bootstrapping, testing strategy, deployment, and documentation lifecycle.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/shandar/pwp-plugin.git"},{"slug":"prontonlp-plugin","title":"prontonlp-plugin","description":"The ProntoNLP Plugin is an advanced financial analysis toolkit designed to transform complex financial data and earnings calls into structured, actionable intelligence. Powered by ProntoNLP data, the plugin enables investors, analysts, and researchers to generate comprehensive analytical reports on individual companies, broad market sectors, overarching market movements, or specific thematic topics—tracking sentiment trends, investment scores, and critical market narratives with ease.\n\nKey Features\n\nCompany Intelligence (pronto-company-intelligence): Conduct deep dives into specific companies. Track quarter-over-quarter shifts in management and analyst sentiment, benchmark stock performance, and identify predictive investment scores.\n\nSector & Industry Intelligence (pronto-sector-intelligence): Analyze entire industries simultaneously to rank companies, spot divergence signals (e.g., high sentiment + weak stock = potential buy), and identify dominant macroeconomic themes.\n\nComparative Analysis (pronto-compare): Compare anywhere from 2 to 5 companies or sectors head-to-head. Generate a unified side-by-side scoring across risk, sentiment, and performance dimensions to find the absolute leader.\n\nMarket Pulse Dashboard (pronto-marketpulse): Get a fundamentals-driven view of broad market activity. Track the biggest sentiment shifts, top movers, and most bullish companies across custom timeframes.\n\nThematic Topic Research (pronto-topic-research): Execute qualitative keyword research across thousands of financial documents. Synthesize trends, related sectors, and verbatim executive quotes into a structured narrative HTML report.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/ProntoNLP/ProntoNLP-Plugin.git"},{"slug":"project-manager-vibes","title":"project-manager-vibes","description":"PM Vibe is a free skill for Claude Code that adds project management and bug prevention for vibe coders. It builds structured specs before writing code, runs security scans after every build, auto-saves progress, and manages design tokens. All activated with pm on. 12 modules hosted with nuggieai.com","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/christinezhang5/project-manager-nuggieai-vibes-skill.git"},{"slug":"product-review-expert","title":"product-review-expert","description":"product-review-expert is a Claude Code plugin for structured product and UX review. It is intended to help users analyze product requirement documents, workflow descriptions, page plans, and interaction designs before implementation or release.\n\nThe plugin focuses on surfacing concrete quality issues rather than generating generic feedback. Its review scope includes workflow completeness, state-transition coverage, exception and edge-case handling, role and permission consistency, rule conflicts, action-path friction, missing feedback, and weak error-state design.\n\nThe repository currently includes:\n\n3 skills:\nproduct-audit\nlogic-defect-review\ninteraction-review\n1 agent:\nsenior-product-reviewer\n1 local hook:\na Stop hook that runs a bundled local script for lightweight normalization/logging purposes\nThe plugin does not depend on external APIs, hosted services, MCP servers, OAuth flows, or user-provided secrets. It does not send user content over the network. The included hook only executes a local bundled script and writes a timestamped diagnostic log under ${CLAUDE_PLUGIN_DATA}. It does not modify user repository files or perform analytics collection.\n\nThe plugin is documented with:\n\na public GitHub repository\na manifest in .claude-plugin/plugin.json\ninstallation and usage instructions in README.md\nsemantic versioning and changelog\nlicense information\nMarketplace-facing usage is documented with namespaced commands such as:\n\n/product-review-expert:product-audit\n/product-review-expert:logic-defect-review\n/product-review-expert:interaction-review\n/product-review-expert:senior-product-reviewer\nThe intended users are product managers, designers, business analysts, and teams reviewing process-heavy or stateful product designs such as approval systems, admin panels, internal tools, B2B workflows, and e-commerce flows.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/YuSec2021/product-review-expert.git"},{"slug":"product-discovery","title":"product-discovery","description":"Run a product initiative as a tracked journey — not a toolbox of one-shot answers. Persistent state across sessions: living PRD, evidence-typed hypotheses (REAL/SYNTHETIC/INFERRED), decision log. From one-sentence problem to PRD, AB test analysis, and GTM rollout to existing users — 19 configurable steps.\n\nKey features:\n• One-sentence start — describe your problem, get hypotheses + research plan in 60 seconds\n• 19-step pipeline across 3 phases: Problem Research → Solution Design + Validation → Launch\n• Living PRD that builds incrementally as you progress\n• Evidence typing (REAL/SYNTHETIC/INFERRED with confidence 0.0-1.0)\n• Cross-session continuity — resume exactly where you stopped\n• Personal context that grows: pm-profile.md, .product-corrections.md, .initiatives-digest.md\n• Tracker integration: push tickets to Jira /Linear / GitHub Issues via MCP\n• 5 pipeline templates: Quick / Full / Problem-only / Solution-only / Custom\n\nFor PMs working on a real initiative who want a structured path from problem to launch — not just one-shot AI answers.\n\nRequires Claude Code desktop or CLI (web version not supported — needs persistent local state).","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/lenar-amirov/product-pipeline-public.git"},{"slug":"product-dev-agent","title":"product-dev-agent","description":"An opinionated product-development loop: a planning model designs the work, an implementation model executes it via outside-in BDD + unit TDD, with a 3-pass critical review on both the plan and the code, and a Keep/Change/Try retrospective that folds learnings back into the rules.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/xavierbriand/product-dev-agent.git"},{"slug":"preserve-session","title":"preserve-session","description":"Preserves Claude Code session history across project directory renames, moves, and copies by assigning each project a path-independent UUID and maintaining a global registry mapping UUID to current path.\n\nAddresses a long-standing pain point: --resume and --continue silently lose session history when paths change. Closes anthropics/claude-code#5768; also tracked in #38089, #28745, #26766, #27883.\n\nCommands: /preserve-session:fix recovers sessions after rename/move (with copy detection and automatic .jsonl merge on destination collision), /preserve-session:inherit copies session history from another registered project, /preserve-session:doctor diagnoses hash/registry state and slug collisions including non-ASCII path issues on macOS, /preserve-session:uninstall removes all preserve-session data with a preview-then-confirm safety pattern. A SessionStart hook auto-initializes .claude/hash.txt on first run — zero manual setup.\n\n  Demo: https://www.youtube.com/watch?v=lWMrzASG7_o\n\n  Quality signals: 24 verified test scenarios (rename, move, copy, slug collision, non-ASCII paths, registry corruption recovery, full --resume E2E). Atomic registry writes via fcntl.LOCK_EX + tempfile.mkstemp + os.replace. NFC slug normalization to match Claude Code's path encoding on macOS. Python heredoc isolation against shell injection. Symlink-safe uninstall (isinstance/dict guards).\n  Shared helpers in hooks/common.sh with cross-platform Python discovery and uuidgen fallback.\n\n  Full design rationale, security notes, and test plan are in two related submissions:\n  • https://github.com/anthropics/claude-code/pull/39148 (initial bundled-plugin attempt; comprehensive design + 24-item test plan with collapsible sections)\n  • https://github.com/anthropics/claude-plugins-official/pull/1347 (auto-closed by submission bot; retained as reference for the proposed marketplace.json entry)\n\n  Plugin location: lives at plugins/preserve-session/ inside the wonbywondev/claude-plugins monorepo. The marketplace entry should use the git-subdir source format pointing to that subdirectory.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/wonbywondev/claude-plugins.git"},{"slug":"prepkit-product","title":"prepkit-product","description":"Product management skills for Claude Code. Gives Claude structured methodologies for discovery synthesis (JTBD, empathy mapping, journey mapping), user research design    \n  (interview guides, screeners, synthesis plans), opportunity mapping (pursue/monitor/defer decisions), PRD authoring (evidence-linked requirements, Given/When/Then scenarios),\n   prioritization (RICE, MoSCoW, ICE scoring), metrics analysis (North Star, metric trees, HEART framework), engagement design (habit loops, gamification, PLG flows), and      \n  validation (evidence grading, cheapest-next-evidence routing). Each skill enforces evidence quality gates and chains into the next logical product step.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/namht1st/prepkit-product.git"},{"slug":"prepare-delivery","title":"prepare-delivery","description":"Pre-ship quality gates without the ship. Runs the full pre-merge pipeline — deslop (clean AI debris), simplify (de-bloat), agnix (lint agent configs), enhance (audit prompts/skills/hooks/plugins/docs), multi-agent review (audit-project), test-coverage check, delivery validation, sync-docs — and produces a unified pass/fail report. Doesn't push, doesn't create PRs, doesn't merge — just gates. Designed for the workflow where you want all the quality enforcement but the actual ship step happens later (or via a separate /ship invocation, or by hand).","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/agent-sh/prepare-delivery.git"},{"slug":"portfolio-review","title":"portfolio-review","description":"Get the kind of portfolio feedback you’d actually get in a hiring room, but rarely hear.\n\nThis skill breaks down your design portfolio across storytelling, structure, craft, and product thinking. It flags what’s working, what’s hurting you, and what’s quietly disqualifying you.\n\nEvery review includes severity-coded feedback (from minor polish to hiring blockers), plus a clear hiring signal and level calibration.\n\nIf your portfolio isn’t landing interviews, this will tell you why and what to fix first.\n\nExport your results as a clean HTML report to share or track progress over time.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/wonjyou/portfolio-review-skill.git"},{"slug":"popper-probe","title":"popper-probe","description":"A research companion based on Karl Popper's philosophy of falsification that interrogates your hypotheses to find what would prove them wrong. It acts like an adversarial supervisor by sharpening the scope, designing concrete experimentation setup, and referring to literature for state-of-the-art approaches.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/mashathepotato/popper-probe.git"},{"slug":"pm-claude-skills","title":"pm-claude-skills","description":"90 Claude Skills across 14 professions — the largest open-source professional skills library for Claude Code. Covers product management, marketing, engineering, data, design, Figma, legal, finance, HR, sales, operations, research, and more. Each skill is a structured SKILL.md file that teaches Claude how to produce professional-grade outputs for recurring workflows. Install the full library in one command or pick individual profession bundles. 210+ GitHub stars, MIT licence, open for community contributions.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/mohitagw15856/pm-claude-skills.git"},{"slug":"pluginos-figma-operations-platform","title":"pluginos-figma-operations-platform","description":"Run any Figma operation from any LLM agent at ~230 tokens per call instead of ~28,000 — a 15× cost reduction on real design workflows.\nPluginOS bridges your Figma files to Claude, Cursor, Windsurf, or any MCP-compatible agent through 5 focused tools that discover operations on demand. No giant tool schemas bloating context. No raw node dumps confusing your agent.\n\n⚠️ Requires the free PluginOS Bridge plugin running in Figma:\n👉 https://www.figma.com/community/plugin/1626608701431483287\n\nOpen it once before your first agent call, it auto-reconnects after that.\n\n- 5 MCP tools, unlimited operations, discovered dynamically at runtime\n- ~6,600 tokens per multi-step workflow vs ~105,000 with traditional approaches\n- Pre-summarized, structured results: your agent reasons faster with less noise\n- 28 built-in operations: lint, audit, content, components, accessibility, more\n- Custom operations: add a manifest + execute pair, no server changes\n- Multi-file support: connect several Figma files at once\nCompatible with Claude Desktop (one-click DXT install), Claude Code, Cursor, Windsurf, and any MCP client.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/LSDimi/PluginOS.git"},{"slug":"playwright-pro","title":"playwright-pro","description":"Playwright Pro turns your AI coding agent into a senior test automation engineer.\n\nInstead of generating tutorial-quality tests with brittle CSS selectors, Playwright Pro enforces battle-tested patterns from production test suites: role-based locators, web-first assertions, proper fixture architecture, and isolated test design.\n\nIt doesn't reinvent Claude Code — it orchestrates it. /pw:generate uses the built-in Explore subagent for codebase analysis. /pw:migrate delegates to /batch for parallel file conversion. /pw:fix leverages /debug for trace inspection. Every skill builds on what Claude Code already does well.\n\n55 ready-to-use test templates across 11 categories — authentication, CRUD, checkout, search, forms, dashboard, settings, onboarding, notifications, API testing, and accessibility. Each template includes TypeScript and JavaScript variants with parametrizable placeholders. Not generic stubs — real patterns like Stripe iframe handling, GraphQL mutation testing, and RBAC permission flows.\n\n3 specialized agents handle complex planning before a single test is written: test-architect maps your application surface, test-debugger diagnoses flaky failures systematically, and migration-planner creates ordered conversion plans from Cypress or Selenium.\n\n2 MCP integrations (TypeScript) connect to TestRail (8 tools) and BrowserStack (7 tools) for teams that need test management and cross-browser coverage. Both are optional — the plugin works fully without them.\n\nSmart hooks auto-validate every test file you write (catches waitForTimeout, hardcoded URLs, non-web-first assertions) and auto-detect Playwright projects on session start. Zero overhead — they run silently.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"alirezarezvani/claude-skills"},{"slug":"plan-review-skill","title":"plan-review-skill","description":"Multi-perspective plan review from VP Product, VP Engineering, and VP Design viewpoints. Identifies blockers, anti-patterns, conflicts, and regressions before implementation starts.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/wrsmith108/plan-review-skill.git"},{"slug":"plan-review","title":"plan-review","description":"Multi-perspective plan review with VP Product, VP Engineering, and VP Design for implementation plans and architecture proposals.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/wrsmith108/plan-review-skill.git"},{"slug":"person-context","title":"person-context","description":"person-context manages real-world individual relationships in depth — the people that matter most to the user (business partners, key clients, friends, family, and the user themselves). It treats observation as cognitive traits, builds a per-person markdown dictionary, and auto-injects the relevant dictionary into Claude's context whenever the person comes up in a prompt. Unlike persona-switching plugins designed for market research or role-play, person-context targets continuous, longitudinal management of a small number of important real persons.\n\nThree components:\n1. Dictionary templates — a 9-chapter catalog plus custom chapters per person.\n2. Inject hooks — UserPromptSubmit auto-injection so the dictionary is loaded before Claude responds; PreToolUse verification that physically blocks responses where the dictionary was not actually read.\n3. Critical Feedback Patterns — four response rules that co-fire with the dictionary (careful interpretation, dictionary accuracy, pressure-test, obvious-judgment).\n\nKey differentiation vs existing claude-personas / claude-persona / TinyTroupe: real persons + the user (not synthetic personas), an observation-driven update loop (extract -> analyze -> apply), and full Japanese + English documentation.\n\nv0.1.0 released 2026-05-15; v0.1.1 (2026-05-17) adds full English README, template, three sample dictionaries, and CHANGELOG.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/tier1kun/person-context.git"},{"slug":"perf","title":"perf","description":"Structured performance investigation with baselines, profiling, and evidence-backed decisions — run by AI agents inside your editor. `/perf` enforces a rigorous 10-phase methodology derived from recorded real investigation sessions: establish a baseline, identify the breaking point, generate hypotheses, profile, run controlled experiments, measure, document. No random micro-optimizations; every change is baselined, isolated, measured, and either kept or reverted with evidence. Designed to keep the agent honest in a domain where intuition is usually wrong.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/agent-sh/perf.git"},{"slug":"perception-first-design","title":"perception-first-design","description":"Perception-First Design (PFD) — psychology-backed design framework. 5-layer dependency stack derives requirements from cognitive constraints before proposing solutions. Auto-activates on design, marketing, copywriting, and communication tasks.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/skovalik/perception-first-design.git"},{"slug":"pencil-dev-skill","title":"pencil-dev-skill","description":"Teaches AI coding tools to design and edit pencil.dev UI files (.pen format) through the Pencil MCP server. Every read and write goes through the MCP for schema validation. Live screenshots show the design as it looks, and the MCP keeps your changes and the editor in sync while the canvas is open.\n\nRecognises pencil.dev signals automatically: mentions of .pen, .lib.pen, the Pencil canvas, or a design-system folder all activate the skill. Works with Claude Code, Codex, Gemini CLI, Copilot CLI, and Cursor. Supports designing screens from scratch, editing existing canvases, instantiating components from .lib.pen libraries, customising design tokens, and bulk-replacing properties across a project.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/Nisus74/pencil-skill.git"},{"slug":"pdf-forge","title":"pdf-forge","description":"Professional PDF generation with Tailwind CSS. Generates slide presentations (16:9) and A4 documents from any content using a complete design system — typography scales, spacing grids, color palettes, and 15 ready-to-use templates. Renders via Playwright/Chromium. Includes brand customization, anti-AI-sloppy aesthetics (zinc backbone, negative tracking, restrained accents), and an MCP server for Claude Desktop integration.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/syx-labs/pdf-forge.git"},{"slug":"paper-desktop","title":"paper-desktop","description":"Design on a canvas that Claude can read and write to — built on web standards.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-14T00:00:01.915Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/paper-design/agent-plugins.git"}],"total":268,"page":3,"per_page":36,"observed_at":"2026-10-14T00:00:01.915Z","source_counts":{"web":341,"demo":13,"skills":5,"official":315,"community":2283,"knowledge-work":123}}