Content
2.5.22
严肃长文流水线（学术论文/商业评论/行业分析/公众号深度长文）——多 Agent 子代理编排。三角验证（文献/数据/案例）+ M 门（LLM 结构化判定）+ F 失败模式防御 + 数据信任 3 档 + 修订回环 ≤2 轮。使用前需 Phase 0 同意关卡。<2000 字建议直接用主控 LLM。
Automation
agent-teams
Coordinates Claude agent teams via filesystem protocol
AI / ML
classroom
Use when modelling a multi-agent learning session — a team of agents in pedagogical roles (teacher delivers, peer debates, examiner gates, debater brings viewpoints) teaches a comp…
Automation
loop-up
Multi-agent loop that auto-corrects multi-stage work: an orchestrator plans and validates upgrades, an implementer executes changes, a read-only observer logs actions, and an inspe…
Automation
canvas-agent-chat-skill
Call a local or remote Canvas Agent `/chat` API and return either the final answer or a structured `question_required` payload. Use this whenever another agent should delegate work…
Engineering
roll-peer
Cross-agent peer review skill. When a task enters a decision phase (planning, high-risk, ambiguous, destructive), triggers a bidirectional negotiation with another AI agent via a u…
Productivity
fleet-economy
Use when orchestrating any non-trivial multi-part task with subagents or model lanes — builds, audits, refactors, research sweeps, migrations — or when deciding which model tier ha…
Research
cars
CARS (Collaborative Academic Research System) - Modular multi-agent academic research system. Uses Socratic dialogue, literature search, and the chessboard reading method to identi…
Automation
oh-my-hermes
Multi-agent orchestration skills for Hermes inspired by `oh-my-claudecode` and rebuilt on Hermes primitives. Suite covers deep-research, deep-interview, `ralplan` (Planner → Archit…
AI / ML
forge-multi-agent
When to spawn subagents versus work inline with patterns including fan-out, sequential, orchestrator/worker, judge/critic, retry-with-fallback, and pipeline. Covers context budget …
AI / ML
ctx2skill
Faithful reproduction of Ctx2Skill (arXiv:2604.27660v2) — multi-agent self-play framework that autonomously discovers, refines, and selects context-specific skills from complex doc…
Automation
zenflow
Multi-agent workflow engine for Go. Declarative YAML workflows; an LLM coordinator routes events through hub-and-spoke mailboxes. Three CLI verbs (flow / goal / agent) cover determ…
Documentation
academic-paper
Multi-agent academic paper pipeline with 10 modes covering full drafting, planning, outlining, revision, abstract generation, literature review, formatting, citation checks, and di…
Productivity
install-department
Install a department's slice of the company brain via multi-agent probes of connected tools (Notion, HubSpot, Linear, Slack, Granola, Gmail) plus structured methodology interview. …
Research
dash-research-vault
Multi-agent research vault with Evensong-style memory causation, Ebbinghaus decay, and L2 pressure self-evolution. Three-tier memory (public/restricted/internal), trigger-based age…
AI / ML
superthink
IPC-based multi-agent review via Crispy's internal dispatch. Dispatches parallel child sessions (claude + codex by default) through the running Crispy host — no subprocess spawning…
Automation
os-improvement-loop
Coordinates concurrent multi-agent improvement cycles using shared event bus and memory. Each cycle executes, evaluates (KEEP/DISCARD), emits friction events, persists metrics and …
Automation
cto-orchestration
Orchestrates multi-agent development as CTO: dispatches execution and review agents via tmux, drives work from goal docs with watcher monitoring, adversarial review loops, flag gat…
AI / ML
sisyphus-autopsy
Forensic reconstruction of multi-agent sessions: rebuilds each agent's state, observations, and reasoning at every decision point to assess whether calls were justified. Use for de…
Automation
ab-scripting-feature-dev
Creates an AgenticBrowser Scripting DSL orchestration for end-to-end feature development: planning, TDD implementation, testing, and lint review. Outputs docs/scripting-features/fe…
Engineering
overnight-multi-issue-implementation
Runs a multi-agent overnight workflow that converts a cluster of related GitHub issues into stacked, merged PRs by morning. Handles 6–15 issues split across two sequential PRs, usi…
Productivity
clawdi
Cross-agent long-term memory and session history capturing user preferences, coding habits, projects, past bugs, architecture decisions, and conversations across multiple agents. S…
Security
brutal-review
Zero-tolerance multi-agent code review system. Launches parallel reviewers across Security, Architecture, Quality, Performance, and Style with MCP integration and weighted scoring;…
AI / ML
PolyBrain
Multi-agent, multi-model orchestration skill that decomposes an objective into a JSON task plan, runs research and build subtasks in parallel, then synthesizes and verifies a final…
Automation
agent-telegram-free
Defines Telegram communication rules for multi-agent setups: accountId, emoji identifiers, and message formats for main/backend/frontend roles. Routes messages via message tool wit…
AI / ML
build-local-context-builder
Orchestrates a multi-agent team (infra-architect, schema-designer, collector-engineer, processor-engineer, integration-qa) to implement the full local-context-builder pipeline from…
Engineering
pantheon-gap
Runs multi-agent gap analysis on an existing project: maps the codebase then fans specialized probes across completeness, security, tests, architecture, and other dimensions to sur…
Engineering
arib-build
Coordinates a multi-agent engineering effort to deliver a specified goal: decomposes work, dispatches specialists in parallel where possible, integrates results, and verifies befor…
AI / ML
Roundtable AI
{ "project": { "name": "roundtable-ai", "description": "A multi-agent system with a FastAPI backend and React frontend for orchestrating discussions among AI agents.", "repository"…
AI / ML
Supacomputa
This document provides the complete **supercompute-programming framework** for orchestrating multiple AI agents in quantum forest ecosystems, enabling seamless collaboration betwee…
AI / ML
tradingagents-analysis
Multi-agent AI tool for China A-share analysis: 15 specialized agents collaborate on technical, fundamental, sentiment, capital flow, macro, and game-theory research to deliver str…
AI / ML
agent-orchestration
Coordinates multi-agent workflows by delegating tasks to specialist sub-agents while preserving their autonomy in execution. Applies a structured framework that supplies world-buil…
Research
deep-research
Orchestrates multi-agent research workflows: decomposes a topic into parallel subtasks, runs them via non-interactive agent sessions, prioritizes installed skills then MCP tools fo…
Automation
copilot-cli
Run GitHub Copilot CLI sessions for agentic coding, session management, custom agents, skills, plugins, SDK integration, and autonomous loops. Supports interactive or non-interacti…
Engineering
harness-plugin-architecture
Designs, reviews, or refactors multi-agent harness plugin adapters across hosts: add/fix host adapters, align N-host capability parity, define Track A/B, and establish capability i…
Design
brand-name-explore
Multi-persona naming system that launches parallel sub-agents, each following a distinct naming philosophy—Lexicon methodology, Poet, Linguist, Culture Hacker, Futurist—to generate…
Engineering
herdr-collab
Primary skill for coordinated multi-agent code review. Maintains canonical contracts, state transitions, tags, and templates for fixed-revision single reviews and dual-panel Herdr …
Engineering
hermes-forgecode
Delegate coding tasks to Forge (Forgecode) CLI agent. Use when you need to build features, fix bugs, refactor code, perform code reviews, or do architecture research. Triggers: for…
AI / ML
orchestrator-core
Domain-agnostic orchestration kernel for multi-agent systems. Extracts a reusable supervisor layer with context stripping, contract-based dispatch, layered merging, and adversarial…
Automation
multi-agent
Spawn worker Claude panes in the current tmux window and send them messages. Use when the user wants to dispatch work to one or more parallel Claude sessions visible alongside the …
AI / ML
Multi-Agent Systems in ADK
<div class="language-support-tag"> <span class="lst-supported">Supported in ADK</span><span class="lst-python">Python v0.1.0</span><span class="lst-typescript">Typescript v0.2.0</s…
Productivity
overnight-review-client-delivery
Runs an overnight autonomous session to generate a polished client delivery package by morning, using a parallel multi-agent review panel to catch errors before hand-off. Fits boun…
Automation
5dive-cli-extras
Manage fleet-scale agent operations: hire/fire personas, BYO providers, multi-account auth, declarative fleets, CrewAI crews, scheduled tasks, multi-agent loops with verifier gates…
Engineering
deadly-loop
Iterative multi-agent debate protocol that spawns Reviewer, Auditor, and Critic roles to surface and resolve P0 blockers in high-risk changes before merge. Loops fix waves until co…
Automation
milestone-workflow
Orchestrates multi-agent milestone workflows from Execution-stamped GitHub issues: builds dependency tracks, presents run plans for approval, then executes per-issue pipelines with…
Testing
adverse-review
Multi-agent adversarial code review that spawns three reviewer personas (Auditor, Adversary, Pragmatist) on one model, runs cross-examination, and synthesizes a ranked report. Use …
Automation
graphsmith
Builds and hardens multi-agent workflows with scaffolding, architecture review, crash-recovery testing, and coordination rules. Use when users request agents, automation, or pipeli…
Automation
grok-swarms
Use when orchestrating complex multi-agent projects. A Grok 4.3 orchestrator plans with the human then dispatches swarms of optimal Grok executors in isolated TMUX sessions using n…
Productivity
strike-trio-en-v1-3-0
Generates and orchestrates a Strike Trio setup with one PM and two AI dev agents. Handles project initialization—collecting details, scaffolding docs, and transitioning to discover…
Content
inkos
CLI agent for novel writing and creative fiction: generates chapters, continues stories, imitates styles, exports EPUB, detects AIGC. Includes 10 English genre profiles plus Chines…
Research
roundtable
Multi-agent debate council — spawns 3 specialized sub-agents in parallel (Scholar, Engineer, Muse) for Round 1, then optional Round 2 cross-examination to challenge assumptions and…
Engineering
pantheon-custom
Runs configurable multi-agent coding tasks with user-selected adversarial verification models (DeepSeek, Qwen, local Ollama) via the codex CLI. Executes parallel test-driven varian…
Engineering
multi-agent-dev-v2
Provides structured workflows and configuration guidance for multi-agent development scenarios, with optimizations addressing common usage pain points. Use for code generation, deb…
Productivity
discipline
Proof-based orchestration discipline for spec-compliant multi-agent systems. Five layers: DECOMPOSITION, COMPREHENSION, VERIFICATION, ENFORCEMENT, ACCOUNTABILITY. Provides discipli…
Engineering
pg-patch-review
Performs multi-agent PostgreSQL patch review: runs mechanical validation then fans out five parallel critics (architecture, breaking changes, test coverage, style, and reflex) befo…
Automation
multi-agent-en
Generic multi-agent dispatcher that converts the primary agent into a coordinator delegating tasks to five persistent sub-agents. Supports round-robin scheduling, reply-before-spaw…
Engineering
review-council
Multi-agent review that delegates to codex-cli (GPT-5.5 xhigh), gemini-cli (Gemini 3.1 Pro), and four specialist subagents covering security, performance, logic, and regression, th…
AI / ML
F1 Strategist AI
This is an AI-powered F1 race strategy assistant with: - Multi-agent architecture (5 specialized agents) - Hybrid LLM system (Claude Opus/Sonnet/Haiku + Gemini) - OpenF1 API integr…
Automation
goap-agent
Invoke for complex multi-step tasks requiring intelligent planning and multi-agent coordination. Supports task decomposition, dependency mapping, and execution strategies including…
Engineering
run-agent-team
Coordinates multi-agent teams for code-changing issues: architect→reviewer→developer→tester→documenter workflow with dependency-aware sequencing and pre-dissolution checks. Falls b…
Showing the top 60 of 1,133. See the full list →