HomeTrang chủ All agentsMọi agent
The complete subagent roster — 20 specialistsDanh sách subagent đầy đủ — 20 chuyên gia

Every agentMọi agent

20 agents

Skills deploy these specialized subagents, each in its own context with its own tools. Most are advisory — they report and do not mutate plan or code unless tasked. The agents guide explains how they compose (the critique panel); this is the full roster with each one's model and tool count.Skill triển khai các subagent chuyên biệt này, mỗi cái trong context riêng với tool riêng. Phần lớn là tư vấn — báo cáo, không sửa plan/code trừ khi được giao. Hướng dẫn agent giải thích cách chúng kết hợp (hội đồng critique); đây là danh sách đầy đủ kèm model và số tool.

brainstormer

Use this agent when you need to brainstorm software solutions, evaluate architectural approaches, or debate technical decisions before implementation. Examples: - - -

model: opus11 tools

code-reviewer

Comprehensive code review with scout-based edge case detection. Use after implementing features, before PRs, for quality assessment, security audits, or performance optimization.

model: opus12 tools

code-simplifier

Simplifies and refines code for clarity, consistency, and maintainability while preserving all functionality. Focuses on recently modified code unless instructed otherwise.

model: sonnet13 tools

critique-consolidator

Use this agent to merge findings from several independent critique lenses into one ranked, de-duplicated verdict.

model: opus9 tools

debugger

Use this agent when you need to investigate issues, analyze system behavior, diagnose performance problems, examine database structures, collect and analyze logs from servers or CI/CD pipelines, run tests for debugging purposes, or optimize system performance.

model: sonnet15 tools

decision-reconciler

Use this agent as Tier-2 of decision-flip governance: a deep, LLM-judgment reconcile pass over the whole Decision Register that the inline script gate cannot do.

model: opus6 tools

developer

Use this agent to implement a scoped feature, module, or fix in an isolated worktree following TDD. Best for the build role in a parallel team where each developer owns a disjoint set of file globs and must not touch others. Examples:\n\n\n\n

model: sonnet15 tools

docs-manager

Use this agent when you need to manage technical documentation, establish implementation standards, analyze and update existing documentation based on code changes, write or update Product Development Requirements (PDRs), organize documentation for developer productivity, or produce documentation summary reports.

model: haiku15 tools

git-manager

Stage, commit, and push code changes with conventional commits. Use when user says "commit", "push", or finishes a feature/fix.

model: haiku9 tools

independent-revalidator

Use this agent to independently re-derive a load-bearing conclusion from primary evidence BEFORE reading the first-round reasoning, then surface where the two disagree. Deploy on plan decisions, review verdicts, or analyses whose correctness the next step builds on.

model: opus10 tools

journal-writer

Use this agent when:\n- A test suite fails repeatedly despite multiple fix attempts\n- A critical bug is discovered in production or staging\n- An implementation approach proves fundamentally flawed and requires complete redesign\n- External dependencies (APIs, services, libraries) cause blocking issues\n- Performance…

model: haiku13 tools

market-fit-critic

Use this agent as a critique lens that judges whether a product-bearing artifact stands up in its market — a plan, a design, or a strategic decision before it is committed. It examines defensibility, differentiation, and the path to value capture, separate from whether users want it or whether it is buildable.

model: sonnet10 tools

planner

Use this agent when you need to research, analyze, and create comprehensive implementation plans for new features, system architectures, or complex technical solutions.

model: opus15 tools

product-value-critic

Use this agent as a critique lens that judges whether an artifact builds something users actually need — a plan, a design, or a product-bearing decision before it is committed. It examines the job-to-be-done, the value proposition, and the riskiest unproven belief, separate from whether the work is technically sound.

model: opus10 tools

project-manager

Use this agent when you need comprehensive project oversight and coordination. Examples:

model: haiku14 tools

red-teamer

Use this agent to adversarially attack an artifact that is ABOUT TO EXECUTE — a plan before cook, a migration before it runs, a gate config before it ships — hunting the failure mode that the author and the happy path both missed.

model: opus10 tools

researcher

Use this agent to conduct comprehensive research on software development topics — investigating technologies, finding documentation, exploring best practices, gathering information about packages and open-source projects. Synthesizes multiple sources into a ranked, evidence-weighted research report.

model: haiku12 tools

tester

Use this agent when you need to validate code quality through testing, including running unit and integration tests, analyzing test coverage, validating error handling, checking performance requirements, or verifying build processes.

model: haiku15 tools

ui-ux-designer

Use this agent when the user needs UI/UX design work including interface designs, wireframes, design systems, user research, responsive layouts, animations, or design documentation. Examples:\n\n\n\n\n\n\n\n

model: sonnet17 tools

workflow-orchestrator

Use this agent to design a spawn strategy for a task you are about to delegate — decide subagents vs Workflow, group the fan-out by concern, size the sub-count, set the batch-consolidate cadence, and lay out the early-write report paths — WITHOUT spawning anything itself.

model: sonnet7 tools