Claude Code subagents
Subagents are specialised assistants Claude Code hands work to: a code reviewer, a test writer, a researcher. Each runs in its own context with its own instructions and tools. 167 plugins in our directory ship ready-made subagents.
Last checked against Anthropic's documentation on .
How a subagent is defined
A subagent is a Markdown file with YAML frontmatter. Only name and description are required; the body becomes its system prompt. Optional fields include tools, model and skills.
.claude/agents/for one project~/.claude/agents/for all your projects- A plugin's
agents/folder
How to call one
- Let Claude delegate: it picks a subagent when your task matches the subagent's description.
- Ask by name, for example "use the code-reviewer agent on this change".
- Mention it as
@agent-<name>. - Start a whole session as it with
claude --agent <name>.
Plugin subagents ignore the hooks, mcpServers and permissionMode fields, a security limit on what a plugin can set.
Plugins that add subagents (167)
Claude for Financial Advisors
Finance
Ready-to-run workflows for financial advisors in Claude Cowork, drawing live data from the tools their firm already uses through third-party connectors. The plugin holds no client data, and the advisor approves any write back to an external system. Claude gathers, organizes, summarizes, compares and presents information, while the advisor exercises judgment, makes the recommendation and gives the advice. For financial professional use only; not investment, legal, or tax advice.
156 starsUpdated Nov 2026Apache-2.0
Agent SDK Dev
Developer Tools
Development kit for working with the Claude Agent SDK
Apache-2.0
Champion Kit
Developer Tools
Help Claude Code champions run hackathons and office hours, find patterns worth sharing, help teammates climb the steps of AI adoption, and write a shareable AI maturity report.
Apache-2.0
Claude Security
Developer Tools
Deep vulnerability scanning of your own code, run entirely inside your Claude Code session at a chosen effort tier, with every finding challenged before it is reported and the verification tally computed in code. Turns surviving findings into targeted patches, each verified by a panel of agents, that you apply when you choose.
Apache-2.0
Code Modernization
Developer Tools
Guided modernization for any legacy codebase: start with /modernize, get an assessment, an interactive map, the business rules mined from the code, and a plan you approve, then an upgrade, a rewrite or a rebuild with independent proof that the new code behaves like the old. Includes security hardening, a one-page report, specialist agents, and an optional live progress pane (early access).
Apache-2.0
Code Simplifier
Developer Tools
Agent that simplifies and refines code for clarity, consistency, and maintainability while preserving functionality. Focuses on recently modified code.
Apache-2.0
Feature Dev
Developer Tools
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Apache-2.0
Hookify
Developer Tools
Easily create custom hooks to prevent unwanted behaviors by analyzing conversation patterns or from explicit instructions. Define rules via simple markdown files.
Apache-2.0
Math Proof
Data Analysis
Two skills for hard mathematics problems, each ending in a self-contained proof.md: /math-proof:solo has the session work the problem itself in stages with a notes file; /math-proof:siege works on it for hours in rounds of judge and worker sub-agents and says plainly what is and is not proved.
Apache-2.0
Plugin Dev
Developer Tools
Comprehensive toolkit for developing Claude Code plugins. Includes 7 expert skills covering hooks, MCP integration, commands, agents, and best practices. AI-assisted plugin creation and validation.
Apache-2.0
PR Review Toolkit
Developer Tools
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
Apache-2.0
Atomic Agents
By BrainBlend-AI
Comprehensive development workflow for building AI agents with the Atomic Agents framework. Includes specialized agents for schema design, architecture planning, code review, and tool development. Features guided workflows, progressive-disclosure skills, and best practice validation.
6,333 starsUpdated Nov 2026MIT
Salesforce Development
By Salesforce
Build Salesforce apps and agents using these core building blocks: metadata, Apex, deploy/retrieve, security, reporting, and generated installed-versus-available capability discovery.
1,167 starsUpdated Nov 2026Apache-2.0
Qt Development Skills
By Qt Group
Agentic engineering skills for Qt software development — Qt C++/QML code review, QML coding, and Qt C++/QML code documentation.
546 starsUpdated Nov 2026
Zapier
By zapier
Connect 8,000+ apps to your AI workflow. Discover, enable, and execute Zapier actions directly from your client.
460 starsUpdated Nov 2026MIT
Output.ai
By Output.ai
Output.ai workflow development toolkit for Claude Code. Adds 5 specialist agents (planner, builder, debugger, prompt writer, quality reviewer), 40+ slash-command skills covering scaffolding, debugging, evaluation, and credential management, plus a SessionStart hook that auto-loads Output SDK conventions so Claude understands the framework before the first prompt.
445 starsUpdated Nov 2026Apache-2.0
Vercel
By vercel
Vercel deployment platform integration. Manage deployments, check build status, access logs, configure domains, and control your frontend infrastructure directly from Claude Code.
337 starsUpdated Nov 2026
CodeRabbit
By coderabbitai
Your code review partner. CodeRabbit provides external validation using a specialized AI architecture and 40+ integrated static analyzers—offering a different perspective that catches bugs, security vulnerabilities, logic errors, and edge cases. Context-aware analysis via AST parsing and codegraph relationships. Automatically incorporates CLAUDE.md and project coding guidelines into reviews. Useful after writing or modifying code, before commits, when implementing complex or security-sensitive logic, or when a second opinion would increase confidence in the changes. Returns specific findings with suggested fixes that can be applied immediately. Free to use.
218 starsUpdated Nov 2026MIT
Agentforce ADLC
By SalesforceAIResearch
Agentforce Agent Development Life Cycle — author, discover, scaffold, deploy, test, and optimize .agent files
117 starsUpdated Sep 2026
SonarQube
By SonarSource
Automatically enforce SonarQube code quality and security in the agent coding loop — 7,000+ rules, secrets scanning, agentic analysis, and quality gates across 40+ languages. PostToolUse hooks run analysis after every file edit. Pre-tool secrets scanning prevents 450+ patterns from reaching the LLM. Slash commands give on-demand access to quality gate status, coverage, duplication, and dependency risks. Includes SonarQube CLI, MCP Server, skills, hooks, and slash commands.
114 starsUpdated Nov 2026
PostHog
By PostHog
Access PostHog analytics, feature flags, experiments, error tracking, and insights directly from Claude Code.
101 starsUpdated Nov 2026
Nimble
By Nimbleway
Nimble web data toolkit — search, extract, map, crawl the web and work with structured data agents
75 starsUpdated Nov 2026MIT
DataHub Skills
By DataHub
DataHub development and interaction toolkit with connector planning, PR review, catalog search, metadata enrichment, lineage tracing, data quality management, and connection setup skills
46 starsUpdated Sep 2026Apache-2.0
Product Tracking Skills
By Accoil
AI agent skills that make SaaS products data-ready for product analytics — from codebase scan to tracking plan to working instrumentation code.
46 starsUpdated Oct 2026MIT
143 more
- Mercado Pago1 subagent: mp-integration-expert
- Honeycomb2 subagents: honeycomb-investigator, instrumentation-advisor
- Migration to AWS7 subagents: generic-phase-worker-rw, generic-phase-worker-rwx, llm2bedrock-code-analyzer, llm2bedrock-code-rewriter, llm2bedrock-log-ingestor, llm2bedrock-prompt-evaluator and 1 more
- Spotify Ads API1 subagent: spotify-ads-request-builder
- FullStory1 subagent: session-context
- AI Plugins11 subagents: ai-sast-remediation, cicd-posture, configuration-automation, dependency-reviewer, findings-browser, malware-responder and 5 more
- Duende Skills2 subagents: identity-server-specialist, oauth-oidc-specialist
- Adspirer Ads Agent1 subagent: performance-marketing-agent
- Domino Data Lab3 subagents: _domino-debug, _domino-deploy, _domino-setup
- ActiveCampaign2 subagents: data-operations, marketing-planner
- Convex2 subagents: convex-expert, convex-reviewer
- CockroachDB3 subagents: cockroachdb-dba, cockroachdb-developer, cockroachdb-operator
- HubSpot Sales2 subagents: account-research, brief-fetcher
- Altimate Code1 subagent: analyst
- Azure Cosmos DB Assistant1 subagent: cosmosdb-expert
- Chronograph LP5 subagents: chronograph-cashflow-forecast, chronograph-commitment-pacing-planner, chronograph-gp-meeting-prep, chronograph-look-through-exposure-scan, chronograph-lp-analyst
- Render1 subagent: render-assistant
- Rootly3 subagents: deploy-guardian, incident-investigator, retro-analyst
- Windsor AI4 subagents: attribution-analyst, business-data-analyst, marketing-analyst, portfolio-analyst
- AWS Startup Advisor7 subagents: generic-phase-worker-rw, generic-phase-worker-rwx, llm2bedrock-code-analyzer, llm2bedrock-code-rewriter, llm2bedrock-log-ingestor, llm2bedrock-prompt-evaluator and 1 more
- Brand Voice5 subagents: content-generation, conversation-analysis, discover-brand, document-analysis, quality-assurance
- Chronograph GP6 subagents: chronograph-budget-vs-actuals-variance, chronograph-fund-quarterly-review-pack, chronograph-gp-analyst, chronograph-markup-markdown-brief, chronograph-portfolio-company-one-pager, chronograph-tvpi-attribution-by-company
- GC AI1 subagent: gc-ai-analyst
- Maven Bio1 subagent: research-assistant
- Oracle AI Data Platform Workbench Databricks Migrator2 subagents: databricks-notebook-analyzer, migration-reviewer
- Claude Obsidian3 subagents: verifier, wiki-ingest, wiki-lint
- Autoharness2 subagents: curator, reflector
- Universal Modder3 subagents: mod-game-worker, mod-offline-worker, mod-orchestrator
- Claude Blog5 subagents: blog-researcher, blog-reviewer, blog-seo, blog-translator, blog-writer
- Digital Marketing Pro24 subagents: agency-operations, analytics-analyst, brand-guardian, competitive-intel, content-creator, crm-manager and 18 more
- AI Business Skills6 subagents: channel-operator, content-producer, design-producer, mkt-strategist, performance-analyst, personal-brand-builder
- Hypergrok Trading Desk7 subagents: desk-lead, execution-trader, market-analyst, research-analyst, risk-manager, strategist and 1 more
- Antigravity For Claude Code1 subagent: antigravity-delegate
- Agentops4 subagents: bulk-reader, code-reviewer, code-writer, researcher
- Showtime10 subagents: brand-designer, creative-director, critic, editor, motion-designer, researcher and 4 more
- Lattice1 subagent: verifier
- Sd0x Harness16 subagents: architecture-designer, brief-writer, code-simplifier, codex-architect, codex-implementer, contract-neutral-reviewer and 10 more
- Great Cto73 subagents: accounting-reviewer, adtech-privacy-reviewer, ai-eval-engineer, ai-prompt-architect, ai-security-reviewer, api-platform-reviewer and 67 more
- Craftcms Claude Skills6 subagents: craft-code-reviewer, craft-code-reviewer-deep, craft-debugger, craft-feature-builder, craft-planner, craft-site-builder
- Claude Homelab1 subagent: notebooklm-specialist
- Craft27 subagents: alchemist, become-researcher, chunk-validator, claims-auditor, conductor, creative-analyzer and 21 more
- Platform Skills1 subagent: platform-bulk-reader
- Markifact MCP1 subagent: performance-marketer
- Session Orchestrator14 subagents: analyst, architect-reviewer, code-implementer, db-specialist, dialectic-deriver, docs-writer and 8 more
- Evolving Lite6 subagents: autoevolve-optimizer, health-monitor, integrity-checker, integrity-fixer, planner, whats-next
- Socialforge5 subagents: carousel-builder, compliance-checker, copy-adapter, image-compositor, quality-reviewer
- Superpowers V3 subagents: builder, reviewer, v-memory
- Antigravity Fullstack Hq10 subagents: architect, backend-specialist, code-reviewer, database-specialist, devops-engineer, documentation-writer and 4 more
- Smart Task Router5 subagents: router-easy, router-hard, router-normal, router-reviewer, router-scout
- Clairvoyance1 subagent: clean-room-alternative
- Design Mastery Claude Code3 subagents: brand-architect, design-master, visual-historian
- Streamsnow4 subagents: sql-review-object, sql-review-optimizer, sql-review-page, sql-review-verifier
- Code Sensei1 subagent: sensei
- Scaffolding13 subagents: analyst, architect, coordinator, debugger, developer, devops and 7 more
- Magician7 subagents: fixer, gatekeeper, guardian, reviewer, sentinel, simplifier and 1 more
- Rubi1 subagent: rubi-pesquisador
- Cadence35 subagents: cad-assumptions-analyzer, cad-assumptions-analyzer-high, cad-assumptions-analyzer-low, cad-assumptions-analyzer-max, cad-assumptions-analyzer-medium, cad-executor and 29 more
- Iceberg2 subagents: iceberg-edit, iceberg-score
- Rappi Plugin Claude Openclaw1 subagent: rappi-agent
- Elnora Merit Aktiva3 subagents: merit-bookkeeper, merit-company-lookup, merit-invoice-creator
- Masterplan8 subagents: mp-adversarial-reviewer, mp-alignment-auditor, mp-intent-assessor, mp-intent-critic, mp-plan-reviewer, mp-planner and 2 more
- Skills6 subagents: code-reviewer, completion-judge, db-migrator, debugger, deploy-checker, perf-monitor
- Elnora Linear5 subagents: linear-issue-creator, linear-issue-reviewer, linear-issue-updater, linear-state-curator, linear-url-to-issues
- River Review1 subagent: river-review
- Jig3 subagents: architect, implementer, reviewer
- Objection2 subagents: accuser, defender
- Task Pipeline1 subagent: code-reviewer
- Agy Plugin Cc1 subagent: agy-rescue
- Fondue5 subagents: role-high, role-low, role-max, role-medium, role-xhigh
- Luciazero1 subagent: lucia-reviewer
- Pixeltable Skill2 subagents: pixeltable-debugger, pixeltable-pipeline-architect
- Version Sentinel1 subagent: version-reviewer.agent
- Youtube Agent Pro6 subagents: README, yt-analyst, yt-editor, yt-packager, yt-producer, yt-strategist
- Boss Skill6 subagents: advocate, builder, builder-deep, errand, fable-advisor, inspector
- Brooks Agent Team3 subagents: copilot, language-lawyer, tester
- Devcycle6 subagents: history-inspector, implementer, on-device-driver, red-team-reviewer, spec-reviewer, task-reviewer
- Heimdall16 subagents: architect, coder, database-architect, design, docs-writer, fixer and 10 more
- Hyperclaude7 subagents: documenter, fixer, implementer, planner, researcher, reviewer and 1 more
- Loanword1 subagent: card-builder
- QA Cube3 subagents: qa-analyst, qa-automator, qa-manual
- Sonmat3 subagents: sonmat-scribe, sonmat-witness, sonmat-worker
- Super Claudio2 subagents: ad-creator, content-pipeline
- Finolabs Vibecode3 subagents: navigator, reviewer, security-reviewer
- FRAME2 subagents: researcher, validator
- Gaia10 subagents: README, cloud-troubleshooter, developer, gaia-operator, gaia-orchestrator, gaia-planner and 4 more
- Power Platform Skills8 subagents: agente-arquitetura, agente-automate, agente-canvas, agente-mockups, agente-pesquisa, agente-prototipo and 2 more
- Throughline5 subagents: build-reviewer, design-reviewer, requirements-reviewer, security-reviewer, ux-reviewer
- Agent Build14 subagents: answer-question, builder, completeness, fixer, gate-warden, hand-test and 8 more
- Agentlint Plugin3 subagents: doctor, fix, security-audit
- Brigade9 subagents: cook, cook-design, dev-teammate, dev-teammate-low, dev-teammate-medium, manager and 3 more
- Exo2 subagents: builder, researcher
- GreyMatter Engine5 subagents: README, anesidora, narcissus, nostromo, sulaco
- Larouex Fullstack Plugin12 subagents: accessibility-compliance-agent, authentication-agent, azure-serverless-agent, code-review-agent, content-seo-agent, devops-azure-agent and 6 more
- LLM Orchestrator5 subagents: orch-debugger, orch-explorer, orch-implementer, orch-researcher, orch-spec-reviewer
- Maddog11 subagents: executor-fast, executor-fast-read, executor-judge, executor-lead, executor-smart, product-be and 5 more
- Maude For Claude1 subagent: maude
- Omni Agent1 subagent: worker
- Onewave Toolkit12 subagents: architect, code-reviewer, content-strategist, copywriter, deal-coach, debugger and 6 more
- Phaser4 subagents: implementer, reviewer, scrutinizer, spec-advisor
- Probe7 subagents: alternative-viewpoints, challenge-assumptions, clarify-thinking, evidence-basis, implications-consequences, question-the-question and 1 more
- Skills3 subagents: evaluator, planner, representative
- Swe Workbench23 subagents: accessibility-auditor, alignment-assessor, architect, auditor, code-impl, conflict-resolver and 17 more
- Vibe Force12 subagents: sf-apex-engineer, sf-deploy-engineer, sf-integration-engineer, sf-lwc-engineer, sf-metadata-engineer, sf-orchestrator and 6 more
- 10000x Engineer4 subagents: executor, grunt, reviewer, scout
- Atelier5 subagents: architect, builder, design-scout, learner, summarizer
- Business Crew18 subagents: analista-datos, analista-thompson, ceo-bezos, cfo-precios, cmo-godin, coo-graham and 12 more
- Claude Kit7 subagents: auditor, claim-checker, debugger, gatherer, plan-critic, researcher and 1 more
- Claude Konsey2 subagents: konsey-baskani, konsey-uyesi
- Claude Reposec5 subagents: deps-agent, git-history-agent, pii-agent, secrets-agent, vulns-agent
- Dhpk32 subagents: INDEX, agy-worker, architect, code-reviewer, codex-reasoner, codex-reviewer and 26 more
- Dispatch13 subagents: brainstormer, solver-crypto, solver-for, solver-jail, solver-llm, solver-misc and 7 more
- Forge Plugins27 subagents: builder, builder-pro, devops, dreamer, evaluator, gate-checker and 21 more
- Found Issues2 subagents: found-issues-fixer, found-issues-sweeper
- Game Studios Plugin51 subagents: accessibility-specialist, ai-programmer, analytics-engineer, art-director, audio-director, community-manager and 45 more
- Icm Spec1 subagent: spec-drift-auditor
- Kinan Orchestre6 subagents: courrier, devis, expert, kinan, lecture, production
- LibreGEO Claude Code5 subagents: geo-ai-visibility, geo-content, geo-platform-analysis, geo-schema, geo-technical
- Logo Design1 subagent: logo-critic
- Metica SDK Agents4 subagents: contracts, unity-compat-checker, unity-integrator, unity-validator
- My Writing Companion16 subagents: beta-reader, blind-reader, brainstormer, character-sim, cold-reader, continuity-checker and 10 more
- Natural Intelligence4 subagents: builder, implementer, investigator, reviewer
- Omc Slim4 subagents: explorer, librarian, oracle, tracer
- Skills1 subagent: eval-grader
- Software Project Governance15 subagents: analyst, architect, code-reviewer, coordinator, design-reviewer, developer and 9 more
- Solo Sdlc6 subagents: product-critic, qa-logic, qa-ui, security-reviewer, solution-architect, tech-lead-reviewer
- Spring Fleet3 subagents: fleet-explorer, impact-analyzer, log-correlator
- Stanford Oval Storm Skills3 subagents: storm-expert, storm-researcher, storm-writer
- Agent Tiers7 subagents: advisor, boss, codex-read, codex-write, reviewer, sublead and 1 more
- AgenticSeal5 subagents: crafter, framer, proofer, sealer, tracer
- AI Overmind1 subagent: splinter-twin
- Astromesh Leia5 subagents: leia-architect, leia-doctor, leia-interpreter, leia-operator, leia-tester
- Cortex2 subagents: cortex-auditor, cortex-role-reviewer
- Dev Crew41 subagents: adversarial-reviewer, api-attacker, api-contract-reviewer, architect, attack-scenario, auth-attacker and 35 more
- Drupal Dev Claude Code1 subagent: drupal-reviewer
- Keel1 subagent: pre-mortem-review
- Mellions Coxen1 subagent: mellions
- Mozart Orchestration44 subagents: AUDIT, CONTEXT-BUDGET, COUNTERPOINT, DELIVER, DIAGNOSE, EVAL and 38 more
- Omarchy Ship3 subagents: omarchy-coverage-reporter, omarchy-plugin-architect, omarchy-submission-auditor
- Quarry18 subagents: ai-agent-exploit, browser-verifier, chain-analyst, engagement-planner, exploit-researcher, finding-checker and 12 more
- Securitymaxxing3 subagents: mapper, reviewer, verifier
- Sharaqi Workflow3 subagents: implementer, planner, reviewer
- Tradefloor23 subagents: bear-researcher, bull-researcher, catalyst-analyst, crowd-persona, desk-head, fundamentals-analyst and 17 more
- UI Evaluator9 subagents: accessibility-auditor, code-reviewer, design-critic, direction-designer, finding-verifier, fix-reviewer and 3 more
Read from each plugin's repository on GitHub, checked November 16, 2026. Stars show interest, not quality: open a plugin to see exactly what it installs.
Questions
What is a Claude Code subagent?
A subagent is a specialised assistant that Claude Code hands a task to. Each one works in its own context window, with its own system prompt, tool access and permissions, so a long search or review does not fill up your main conversation.
How do I use a subagent?
Claude delegates to a subagent on its own when a task matches the subagent's description. You can also ask for one by name, mention it as @agent-<name>, or start a whole session as that agent with claude --agent <name>.
How does a plugin add subagents?
A plugin puts subagent files in its agents/ folder. For security, subagents from plugins ignore the hooks, mcpServers and permissionMode fields that your own subagents can use.
More parts a plugin can add: hooks, MCP servers and slash commands.
Sources: Subagents, Plugins reference.