Agents

gorillabiscuit/AI-assisted-project-starter

Agent Claude Code

Cold-context adversarial reviewer for the pre-PR meta-check (AGENTS.md §7 step 3). Reviews the branch diff against AGENTS.md, ADRs, and per-package rules with a deliberate bias toward suspicion. Spawn fresh — it must not see the implementation conversation.

noah-hrbth/agent-tool-sync

Agent Claude Code

Read-only review of a single tool adapter file in internal/tools/ for compliance with agentsync conventions. Invoke after writing or substantially modifying an adapter, or before merging an adapter PR. Pass the adapter file path as the only input.

paruff/uFawkesAI

Agent Claude Code

Build, test, verify, and review an already planned feature. Operates on a feature branch off trunk; prepares a PR but does not merge.

jeromekneip/ai-team-starterkit

Agent Claude Code

Design checklist (from the scout skill — apply before shipping a new agent).

thongton11314/agent-coding-template

Agent Claude Code

Use when making any code change — adding features, fixing bugs, refactoring, updating modules, editing configuration, or changing any file in src/, tests/, wiki/, config/, or scripts/. Enforces the full post-change pipeline: wiki update → sync gate → tests → README → git push.

luabagg/agent-skills

Agent Claude Code

Deep technical advisor for architecture, tradeoffs, plans, risks, and code review. Use before major changes or when implementation choices are unclear.

wonjun3991/study

Agent Claude Code

Feature implementation orchestrator. Invoke with Jira ticket URL (e.g., TMS-320) to implement complete features. Reads requirements from Jira, coordinates all specialist agents, and delivers production-ready code following all project standards.

Indium-AI-Labs/indium-agentkit

Agent Claude Code

Audit data pipelines, models, quality, and governance read-only.

katalalab/katala-os

Agent Claude Code

Delegate a task to Antigravity CLI (agy) running headless in the current working directory. Best for broad-context review, alternate critique, second opinions, large-document/source comparison, and the Google-model perspective. Use for independent verification and consensus, or as the Google brain in a fan-out.…

stephanemalho/multi-harness-agent-template

Agent Claude Code

Check whether the current branch follows the agent branch policy and is up to date with origin/main.

ivanshtokov/copilot-kit

Agent Claude Code

Multi-agent coordination and task orchestration. Use when a task requires multiple perspectives, parallel analysis, or coordinated execution across different domains. Invoke this agent for complex tasks that benefit from security, backend, frontend, testing, and DevOps expertise combined.

nchdatta/projectkit

Agent Claude Code

Owns src/services, src/hooks/queries and src/hooks/mutations. Use to expose a finished API endpoint to the client — entity types, service functions, TanStack Query hooks and keys.

snoodleboot-io/prompticorn

Agent Claude Code

Purpose: Design machine learning pipelines, model training, deployment, and inference systems with specialized expertise When to Use: Working on mlai tasks.

shahidshabbir-se/my-pi-setup

Agent Claude Code

Intelligent task orchestrator that analyzes requests and delegates to specialized agents. Use as the default agent for complex tasks requiring planning, implementation, and review.

Ivan-Linnik/ai-templater

Agent Claude Code

Use for backend architecture analysis, solution design, and impact analysis. Covers API design, service boundaries, data models, auth, and performance.

theogravity/bun-elysiajs-starter-turbo-monorepo

Agent Claude Code

Expert code reviewer specializing in code quality, security vulnerabilities, and best practices across multiple languages. Masters static analysis, design patterns, and performance optimization with focus on maintainability and technical debt reduction.

ulpi-io/agent-library

Agent Claude Code

Expert Express.js developer specializing in middleware architecture, RESTful APIs, queue systems with Bull, production-ready Node.js applications with Pino logging, and enterprise-grade server-side development.

engindemirog/ai-native-engineering-workspace

Agent Claude Code

Independent, read-only code reviewer. Use for the REVIEW step of any workflow — reviews a change set against its spec. Must never write or modify files.

chipi/agentic-ai-homelab

Agent Claude Code

Active triager — normalizes a raw bug issue into a fix-ready L1 problem or rejects it. First pass on every bug-labeled issue; re-entered on harness kick-back with the failed attempt as evidence. Never fixes, never localizes on first pass.

goemktg/Prompt_Template

Agent Claude Code

System complexity, safety, and resource analysis specialist. Performs deep research on constraints, failure modes, and engineering trade-offs. Triggers: system constraints, failure analysis, complexity research, resource limits, safety research.

bertughas123/multi-agent-coding-starter

Agent Claude Code

Read-only repository explorer that gathers exact paths and evidence before implementation begins.

careerchain-ys/stdd

Agent Claude Code

A specialist role for reviewing tests during the Red stage of Spec and Test Driven Development, where tests are written before the implementation. It checks whether tests actually verify the stated requirements.

sgaabdu4/abid-agents

Agent Claude Code

Run MCP-only work in an isolated Pi subagent. Activate needed MCP servers, call MCP tools, return only relevant findings/evidence to the parent agent.

GlamgarOnDiscord/claude-saas-blueprint

Agent Claude Code

Ce dossier est réservé aux vrais subagents Claude Code : fichiers Markdown avec frontmatter YAML (name, description, tools, model, memory, hooks).