Subagents with their own prompt and tools. Injected whole, so what you see in the token column is what every session pays.
Agent Claude Code
The agent system contains 119 specialized AI agents organized into 4 tiers. Each agent is a markdown file defining its role, tools, model preference, and behavioral instructions. Agents are spawned by the router via Task() — they never run independently.
alvin-reyes/better-agentic-ide
Agent Claude Code
Verifies a story against its acceptance criteria and runs the verification command. Use before any story is marked Done.
YuDefine/nuxt-supabase-starter
Agent Claude Code needs its repo
A code-review agent for Nuxt 4, Vue 3, TypeScript, and Supabase projects. Code review is the practice of checking proposed changes for problems before they are merged or committed.
Agent Claude Code
You are the Synthesis subagent for calibration. Your job is to synthesize findings into falsifiable decisions while preserving dissent.
yaukwan/my-claude-code-workflow
Agent Claude Code
Universal implementation and progress tracking expert using artisan agent pattern. Analyzes task specifications, implements solutions following project patterns, validates against acceptance criteria, then automatically updates completion status. Use PROACTIVELY for systematic task implementation with integrated…
ErkanBarin/playwright-agent-mcp-starter
Agent Claude Code
Explores a website, maps user journeys, generates Playwright tests, and detects missing coverage.
stanleykao72/claude-code-spec-workflow-odoo
Agent Claude Code
Design validation specialist for Odoo ERP modules. Validates technical design documents against Odoo architecture patterns, ERP integration standards, and module development best practices.
Agent Claude Code
A specification for writing core-agent cards: files that define an agent’s role, project rules, allowed knowledge sources, recovery steps, and links to reusable skills.
Agent Claude Code
A testing-only workflow that turns a bug report into one test that should fail against the current code. It does not modify the implementation and requires the failure to be demonstrated.
Agent Claude Code
Use to find code and answer questions about it: where something is defined or used, which files match a pattern or naming convention, what a piece of code does. Name the breadth in the request: quick, medium, or very thorough. Read-only. Never for judgment, review, design, or changing files.
kylemoschetto/vibe-md-templates
Agent Claude Code
Technical/Solution Architect agent that reviews backlog, advises on feature integration, evaluates technical approaches, and coordinates with Scribe for backlog management. Routes to specialist sub-agents based on request type.
Agent Claude Code
Designs feature architectures by analyzing existing codebase patterns and conventions, then providing implementation blueprints with specific files to create or modify, component designs, data flows, and a build sequence.
Agent Claude Code
Master Go 1.21+ with modern patterns, advanced concurrency, performance optimization, and production-ready microservices. Expert in the latest Go ecosystem including generics, workspaces, and cutting-edge frameworks. Use PROACTIVELY for Go development, architecture design, or performance optimization.
Agent Claude Code
An operator for carrying out explicitly requested tasks on a Windows computer through visible application actions. It requires checking the available system information and breaking the task into small, verifiable steps.
Agent Claude Code
Finds bugs, missed requirements, and quality issues in plan implementations. Spawned after all tasks pass review for a final sweep.
Agent Claude Code
Review a change against the history of the code it touches. Use git blame and log to catch changes that contradict why the surrounding code exists — reintroducing a fixed bug, removing a guard that was added on purpose, or undoing a deliberate decision. Reports only high-confidence findings backed by a specific past…
Agent Claude Code
Measures whether the same functionality can be delivered with less - reuse that was missed, indirection that forwards without deciding, generality nothing consumes, control flow with a flatter equivalent, and response payloads that spend more of the calling agent's context than their information is worth.
SteveGJones/ai-first-sdlc-practices
Agent Claude Code needs its repo
Orchestrates AI-First SDLC v3 setup by discovering project needs and assembling the right team.
Agent Claude Code
Comprehensive UI/UX testing specialist using Playwright for enterprise-grade user experience validation.
cenconq25/claude-code-app-studio
Agent Claude Code
Owns Cloudflare platform usage end to end: Workers, Durable Objects, D1, R2, KV, Queues, Workflows, Vectorize, Hyperdrive, Workers AI, Agents SDK, Pages, Tunnel, DNS, WAF, and Wrangler. Engage when the backend (or any edge surface) runs on Cloudflare, when wrangler.jsonc needs designing, or when bindings, limits, and…
Agent Claude Code
Use this agent when the user needs deep investigation on a topic, technology, library, API, market data source, academic papers (arXiv/SSRN), or any subject requiring web research.
fieldsphere/cursor-team-marketplace-template
Agent Claude Code
Technical devil's advocate that challenges feature plans, strategies, and implementation approaches when the user launches a request in plan mode. This subagent should be used by the Ask Clarification Tool before creating a plan when planning features, designing systems, or proposing solutions. It should be triggered…
Agent Claude Code
Generic workflow runner for the wiki skill. Executes exactly one documentation phase (repo-scan, toc-design, doc-write, validate-docs, doc-summary, or incremental-sync) from its phase spec and produces the declared outputs.
Agent Claude Code
Code review specialist. Use for correctness, security, regression, and meaningful test-gap review.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: