Agent Plugins Marketplace

Recently added

Plugins grouped by the day they were added to the directory, newest first.

Sep 11, 2026

  1. carrel-agent1stars

    Agentic file librarian for carrel desks: /index, /doctor, /catalog, /completion, a file-librarian agent, an agent-workflows skill, the carrel MCP server (14 tools + carrel:// resources), and a PostToolUse hook that re-indexes files Claude writes.

    Claude Code1 skill1 MCP server
  2. pb-skills2stars

    Skills estándar de Desarrollos Palo Blanco: capa base senior-dev que delega en transversales, stack y producto.

    Claude Code7 skills
  3. pakt0stars

    Draft, review, activate, and use proof-checked trading policies through Pakt.

    Claude CodeAgent Plugins1 skill1 MCP server
  4. superstorm0stars

    Large-goal autonomous pipeline: decompose into modules, front-load decisions via brainstorming, then autonomously design → closed-loop validate → plan → reverse-review → orchestrate → concurrently execute with anti-fake-completion supervision. Global command /superstorm. Also ships /cross-exam (completion audit), /product-review (product-thinking critique) and /keep-code-simple (code-simplicity review); the latter two are advice only.

    Claude Code4 skills
  5. meta-skill0stars

    Meta-skill generator: analyzes target projects and generates project-specific skills, commands, and state machine configurations. Replaces all 6 specialized skills with a unified /bootstrap + /run workflow.

    Claude Code2 skills1 MCP server
  6. grillstorm0stars

    Orchestrates official Matt Pocock design skills, then delivers with reverse-grill gates, supervised concurrency, and post-delivery probe critique.

    Claude Code1 skill
  7. kb-compiler2stars

    Documentation standard for a knowledge base (KB) — section structure, exact Word formatting, KB-XXX naming conventions, and a master-index spreadsheet — with a read-only audit script, a bootstrap script and a test suite. English is the canonical language; a base may be authored in any language confirmed at setup.

    Claude Code1 skill
  8. resell-pro0stars

    Research observed Vinted market data with Resell Pro.

    Claude CodeAgent Plugins3 skills1 MCP server
  9. career-agents0stars

    Career agents for anyone: sync LinkedIn, CV and personal website from one source of truth, review and score all three, find and apply to jobs, track the pipeline, prep and mock every interview stage, write posts with visuals, grow reach, go multi-platform and multilingual, review offers, see it all on a dashboard, and run it unattended on a schedule with email, calendar and digest integrations. Automation modes draft/assisted/auto: LinkedIn MCP for reads and messages, a browser operator agent for profile edits, posts and applications.

    Claude Code30 skills1 MCP server
  10. util-review0stars

    Review Agent Skills, hooks, AGENTS.md/CLAUDE.md files, scripts, MCP configs, or workflow configurations for design flaws, unclosed loops, stale references, and risks.

    Claude Code1 skill
  11. strangler-fig0stars

    Clean-room legacy reimplementation — distill to requirements, rebuild greenfield behind a context firewall, audit each crossing for leaked legacy structure (taint/provenance), verify parity, optionally cut over.

    Claude Code1 skill
  12. session-tracker0stars

    Local MCP server that tracks your Claude Code / Cursor / Codex agent sessions — list, search, label, and inspect changes across every session. Lexical (FTS5) search by default; semantic search is opt-in.

    Claude Code1 MCP server
  13. schema-review0stars

    Review database schemas and in-code data structures for correctness, integrity, performance, scalability, and design quality.

    Claude Code1 skill
  14. retrieval-review0stars

    Audit a retrieval / vector-index pipeline for structural health across seven layered axes — eval foundation, corpus & chunking, embedding geometry, index & ANN fidelity, retrieval composition, rank fusion, and reranking — and flag the defects that silently return wrong or missing context before they reach the generator.

    Claude Code1 skill
  15. ontology-review0stars

    Audit a knowledge graph or ontology for structural health across seven axes — orthogonality, granularity, taxonomic hygiene, identity & rigidity, relationship semantics, competency questions, and inference safety — and flag defects before they corrupt inference.

    Claude Code1 skill
  16. gdpr-review0stars

    Audit a repo, design doc, IaC, or data-flow map against GDPR — lawful basis, consent, data subject rights, retention/erasure, transfers, processors, privacy by design — with article-cited, evidence-backed findings, or assess organizational GDPR readiness. Analysis aid, not legal advice.

    Claude Code1 skill
  17. foundry0stars

    DECOMMISSIONED for public use. The original autonomous-fix loop (Worker/Integrator over isolated git worktrees) has been retired; the skill is now a compatibility shim that routes into the maintainer's private Sven Unit runtime, not distributed in this repo. Not functional for other marketplace installs.

    Claude Code1 skill
  18. diagnose-queries0stars

    Diagnose slow database queries across Postgres, the ORM layer (N+1), and Snowflake/dbt — routes to five modes (known query, locate offender, live incident, proactive audit, regression) and produces a root-caused, verified fix.

    Claude Code1 skill
  19. debut0stars

    Pre-public open-source readiness audit — secrets/PII, licensing, community-health files, tests/CI, deps — scored with fix commands.

    Claude Code1 skill
  20. curriculum0stars

    Generate an adaptive learning curriculum for any topic, with an append-only assessment loop that tunes future modules to the learner.

    Claude Code1 skill
  21. create-skill0stars

    Author a Claude Code (or any host's) Agent Skill end-to-end — scaffold, write against a portable authoring contract, structurally validate, and design-score — in one guided lifecycle. Spec-pure by default; Claude-only affordances are opt-in.

    Claude Code1 skill
  22. config-chunks0stars

    A package manager for agent-instruction guidance — publishes versioned, scored guidance 'chunks', reconciles them into a single bundle, and wires it into CLAUDE.md (@import) and/or AGENTS.md (inlined). Includes /ai-setup onboarding and a chunk-review scoring rubric.

    Claude Code5 skills
  23. compliance-review0stars

    Audit a repo, ADR/PRD, IaC, or live cloud state against SOC 2, HIPAA, or PCI-DSS with structured, evidence-backed findings — or assess organizational readiness for a SOC 2 examination (governance, evidence gaps, phased remediation).

    Claude Code1 skill
  24. bash-gate0stars

    A PreToolUse Bash gate that auto-allows provably-safe commands (so you stop approving the same harmless `mkdir`/`rm` of a build artifact) while still deferring anything dangerous. Deterministic allow-classes by default; an optional LLM arbiter handles the long tail. Ships safe-by-default (no path is auto-allowed until you opt in) and a /bash-gate-add skill to extend it when a prompt slips through.

    Claude Code1 skill
  25. apply0stars

    Browser-driven job application copilot — locate matching postings, score fit, tailor resume/cover letter/screening answers per job to score well on real evaluator engines (Workday HiredScore, Taleo, iCIMS, Ashby, LinkedIn, LLM screeners) with a deterministic keyword-coverage check, drive the application form with a human confirmation gate before every submit, and track the pipeline locally.

    Claude Code1 skill
  26. all-skills0stars

    Meta-plugin that installs every mnox-ai agentic-review skill at once via plugin dependencies. Standalone utility plugins (session-tracker, config-chunks, create-skill, bash-gate) are installed separately. Excludes foundry, which is decommissioned for public use.

    Claude Code
  27. aio0stars

    Agentic Implementation Optimizer — audit, build, or extend agentic AI implementations for reliability, cost, and architecture.

    Claude Code1 skill
  28. diffwitness0stars

    Independent causal verification for code produced by Codex and other coding agents.

    CodexClaude Code1 skill
  29. matrixark-memory215stars

    Time-aware long-term memory for Claude Code, backed by the Rust TemporalStore engine. Automatically ingests each session and injects a ranked ContextPack on every prompt, and exposes explicit recall/remember tools over MCP.

    Claude Code1 MCP server
  30. temporalstore-agent-hooks216stars

    Global Codex lifecycle hooks for MatrixArk context memory backed by TemporalStore.

    Codex
  31. leos-agent2stars

    Adaptive delegation, cost-aware model routing, and deferred workflows across coding harnesses.

    CodexClaude Code7 skills
  32. nereus2stars

    개인 개발 하네스: intake → spec → build(TDD) → review → finish 워크플로, 전문가 에이전트, Baton 컨텍스트 핸드오프

    Claude Code2 MCP servers
  33. credstore2stars

    자격 증명 메타데이터 카탈로그 MCP. 시크릿 값은 저장·반환하지 않고 creds_run으로 마스킹 실행만 한다.

    Claude Code1 MCP server
  34. tribeunal0stars

    Put a decision to a jury on Tribeunal. Connects to the hosted Tribeunal MCP server (https://mcp.tribeunal.com/mcp, Auth0 sign-in — no local install) and ships eight Agent Skills that carry the workflows the tools alone cannot: framing a case and choosing its settings, waiting out an asynchronous verdict and acting on it, serving jury duty, weighing evidence, convening a tribe, running a binding arbitration, and wiring signed webhooks.

    Claude Code9 skills1 MCP server
  35. alive127stars

    Local, user-owned context for Claude Code. Keep your work in files you control.

    Claude Code
  36. interview-research1stars

    Researches a named company's interview process from public web sources and produces an evidence-graded interview-prep report — every predicted question labeled by how well-grounded it is, never a list of generic questions dressed up as company-specific ones.

    CodexClaude Code1 skill
  37. prompt-reverse-engineer2stars

    多模态 Prompt 逆向工程：将文本/图片/视频/剧本作品拆解还原为可复用的专业 Prompt，适配 Midjourney / Stable Diffusion / GPT-4·Claude / DeepSeek / Sora·Runway 等模型格式，附百分制质量评分。

    Codex
  38. context-tree0stars

    Durable, compact context trees and handoffs across Codex tasks

    Codex1 skill1 MCP server
  39. whiting0stars

    Bootstraps repository release discipline: Conventional Commits, semver versioning, changelogs, releases, and audits.

    Codex4 skills
  40. tri-swiss0stars

    A governed red and turquoise Swiss visual system with exact Tailwind tokens and structural component rules.

    Codex
  41. three-axes-framework0stars

    Governs how AI assists developers by calibrating behavior across Mastery, Consequence, and Intent, with thirteen citable Integrity Rules that keep the assistant honest about what it changed and what it verified.

    Codex1 skill
  42. sage-instructor0stars

    Adaptive programming instructor with discovery-first teaching, curricula, and persistent progress tracking.

    Codex1 skill
  43. machine-pilgrim0stars

    Create finished Descent into the Machine illustrations grounded in its canon and visual reference library.

    Codex1 skill
  44. lux-visual-systems0stars

    Art-direct and generate imagery through Lux Solari's Swiss editorial, anime, analogue, and technical visual system.

    Codex1 skill
  45. lux-swiss0stars

    A governed Swiss visual system with exact Tailwind tokens, component rules, and house-mark guidance.

    Codex1 skill
  46. hannah0stars

    F1 strategy engineer for your LLM garage: analyses a repository and hardware to recommend local models.

    Codex1 skill
  47. anime-identity-designer0stars

    Create recognizable anime character identities, portraits, branding boards, and production sheets from real people.

    Codex1 skill
  48. story-flow12stars

    Agentic development workflow covering planning, analysis, implementation, and QA automation

    Claude Code4 MCP servers
  49. junior-flow12stars

    Helps junior developers create technical story markdown from a story tracker

    Claude Code1 MCP server
  50. design-flow12stars

    Agentic design workflow for collaborative product and UI design exploration

    Claude Code