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
Autonomous audit-fix loop aligned with the Anthropic 2026 pattern (code-review plugin). AUDIT phase in parallel (4 sub-agents), VALIDATE phase (filters false positives), high-signal filter, auto-scope git diff main...HEAD. Flags --audit-only and --comment for read-only and inline PR post modes.
Agent Claude Code
Creates the complete design system — tokens, scales, and component patterns. The most important agent in the framework. Spawned by /gdd:spec-phase 1 or /gdd:edit-tokens.
Agent Claude Code
General-purpose software engineer. Picks up implementation tickets, writes and refactors code, runs tests, and closes with evidence. Good default for any ticket without a more specialized agent. Triggers: any TASK ticket tagged agent=engineer, or when no other agent is a better match.
wangjialiang678/research-workflow-skill
Agent Claude Code
A research-focused agent for technical investigations. It prioritises search tools, follows specified output requirements, and saves reports in the project's docs/research/ folder.
Agent Claude Code
GLM advocate for strategic debates - presents alternative perspective via GLM API.
jed1978/ddd-architecture-coach
Agent Claude Code
Implements BC components using TDD workflow. Reads the canonical Phase 3 spec at {coachoutputroot}/{bc}/spec.md plus the .claude/rules/{bc}.md quick-reference card and CLAUDE.md, then follows Red-Green-Refactor per Aggregate vertical slice. Stack-agnostic skeleton — all stack-specific commands (test runner, ORM…
ayeshakhalid192007-dev/graph-engineering-crash-course
Agent Claude Code
A validation agent for this pattern's output.
bhuiyanmobasshir94/claude-standing-orders
Agent Claude Code
Read-only code reviewer. Use after implementation work and before integration, especially on multi-file changes, refactors, and anything touching auth, credentials, migrations, money, PII, or public API shape. Reads the diff and measures it against the project's stated standards, citing file:line for every finding. It…
IncommensurableHubris/fullstack-director
Agent Claude Code
Fullstack Director's blind, read-only OWASP panel reader (skill 07). Each spawn owns exactly ONE area-slice (given in the prompt — classic R1–R4/R5 or the agent-system flipped partition), analyzes only its slice with a neutral evidence-required stance, may run that area's deterministic scanners, and returns findings…
Agent Claude Code
Use for one well-scoped feature, bug fix, test update, or routine refactor after the affected code path and acceptance criteria are understood.
Agent Claude Code
Research agent with enforced epistemic rigor. Classifies all findings as Verified, Inferred, or Guess with cited evidence. Use when investigating code, debugging, exploring unfamiliar systems, or when the user asks to research something. Can persist findings as pkm .ref.md artifacts when instructed.
Wania-Kazmi/claude-code-autonomous-agent-workflow
Agent Claude Code
Test-Driven Development specialist enforcing write-tests-first methodology. Use PROACTIVELY when writing new features, fixing bugs, or refactoring code. Ensures 80%+ test coverage.
Agent Claude Code
You are the [Cx Role Title] for this project. You own [domain description].
Agent Claude Code
Marketing strategy, positioning, copywriting, SEO, CRO, and content creation. Works as a peer to the PM — not a subordinate. Gives the product/market direction while the PM gives the technical direction. Use PROACTIVELY when: working on landing pages, pricing pages, social media posts, email sequences, SEO…
Agent Claude Code
Focused build worker for explicitly delegated Embedded IDE for VS Code rebuild tasks.
Agent Claude Code
Use this agent when the user explicitly requests help with Swift, iOS, or macOS development tasks. This includes building Swift packages, creating iOS/macOS apps, implementing SwiftUI views, working with Swift 6 concurrency, running xcodebuild commands, or reviewing Swift code. Context: User is starting a new iOS…
KeenBrain/awesome-ai-resources
Agent Claude Code
Agent "trust-and-safety" from KeenBrain/awesome-ai-resources, covering :shield: jordan — trust & safety lead, color, persona, expertise and when reviewing documents.
Agent Claude Code
This file is the spec-manager skill-like entrypoint for Codex, OpenCode, and other AGENTS.md-compatible tools. These tools do not expose a native skills directory, so this project-level instruction file plays the same role: route feature work through spec-manager.
aleksandr-chaika/flutter-clean-arch-skills
Agent Claude Code
Flutter Project Manager. Decomposes requirements into structured tasks. Maintains task registry (.tasks/REGISTRY.md) with history. Searches for similar past tasks. Creates test scenarios and Maestro E2E test cases for every task. Supports Asana task URLs as input.
Agent Claude Code
Validate work in tasks-in-testing.md. Full acceptance criteria are in the task block. One item at a time, always sequential. Never fixes bugs, only flags and sends back.
Agent Claude Code
A review agent for checking whether a software project is moving in the right direction at key checkpoints, called stage gates.
Agent Claude CodeCursor
Interactive onboarding that collects user context and configures the PM Operating System — rules, products, and MCPs. Use when user says "onboard", "setup", "PM-OS setup", or "get started".
Agent Claude Code
Cognitive scientist and design theorist. Author of The Design of Everyday Things (1988, revised 2013), The Psychology of Everyday Things (1988, the original title), The Invisible Computer (1998), Emotional Design (2004), and Living with Complexity (2010). Co-founder with Jakob Nielsen of the Nielsen Norman Group…
colinwilliams91/gaitor-orchestrator-cli
Agent Claude Code
Decomposes high-level tasks into parallel sub-tasks and dispatches them to specialised agents. Ensures all agents share the same workspace context and resolves conflicts between agent outputs before merging results.
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: