Multi-Agent agents

1,095 tagged Multi-Agent, measured the same way as everything else here.

Browse within: agent-teams 146Orchestration 112opencode-plugin 110agent 108agentic 48agent-orchestration 45bun 40opencode 40clawdbot 36discord 36discord-bot 36self-hosted 36autonomous-development 34code-quality 34

coder

241

Charpup/brainstorm

Agent

An engineering agent that evaluates technical feasibility and development effort from a coding perspective.

not rated 2 4mo ago A 0 tokens

council-adversary

242

gamgee-ai/council-of-gamgee

Agent

Devil's advocate, red team, pre-mortem analysis. Assumes every plan has a fatal flaw and finds it. Invoke after forming a plan, before committing to a decision, or when something feels too good to be true.

not rated 2 6mo ago A 55 tokens original MIT

council-burt

243

gamgee-ai/council-of-gamgee

Agent

Network structure and social capital analysis. Structural holes, brokerage, network topology, information flow. Invoke for questions about positioning, relationships, market structure, competitive advantage, or when you need to understand who is connected to whom and where the gaps are.

not rated 2 6mo ago A 55 tokens original MIT

council-girard

244

gamgee-ai/council-of-gamgee

Agent

Mimetic theory and social psychology analysis. Desire, rivalry, status games, hidden motivations. Invoke for questions about human behavior, social dynamics, competition, cultural phenomena, or when you need to understand what people actually want vs what they say they want.

not rated 2 6mo ago A 58 tokens original MIT

policy-guardian

247

shihchengwei-lab/separation-and-audit-claude-code

Agent

Policy-layer text reviewer. Reads user-visible content (copy, UI strings, error messages, docs) and returns an evidence-level audit against your policy checklist. Runs mechanism recall on Step 0 automatically.

not rated 2 4mo ago A 44 tokens original MIT

architect

248

jjilli/fable-flow

Agent

Part of fable-flow

Planning agent for the fable-flow pipeline. Delegate to the architect when a task description and scout digests are ready to be turned into an implementation plan with file-disjoint parallel tracks. The architect plans; it never edits files.

not rated 2 1mo ago A 48 tokens original MIT

implementer

249

jjilli/fable-flow

Agent

Part of fable-flow

Implementation agent for the fable-flow pipeline. Delegate one implementer per plan track; each runs in its own isolated git worktree and can safely execute in parallel with other implementers. Also used to apply review fixes on an integration branch.

not rated 2 1mo ago A 51 tokens original MIT

reviewer

250

jjilli/fable-flow

Agent

Part of fable-flow

Adversarial review agent for the fable-flow pipeline. Delegate reviewers (several in parallel, each with a different lens) after tracks are merged into an integration branch, to verify the diff against the plan and requirements before a PR is opened.

not rated 2 1mo ago A 53 tokens original MIT

mumerfarooq-con/asdlc-tools

Agent

Part of asdlc

Pre-delivery inspector for API contract consistency and architectural consistency. Derives the codebase's own conventions and flags deviations. Produces findings only; never edits code.

not rated 2 1mo ago A 39 tokens original MIT

inspector-error-edge

252

mumerfarooq-con/asdlc-tools

Agent

Part of asdlc

Pre-delivery inspector for error-handling completeness and edge-case coverage across the whole codebase. Produces findings only; never edits code.

not rated 2 1mo ago A 33 tokens original MIT

pr-reviewer

253

mumerfarooq-con/asdlc-tools

Agent

Part of asdlc

Reviews a pull request — GitHub PRs via the gh CLI, Bitbucket PRs via the Bitbucket MCP server (PR metadata, description, diff, existing comments), with local git diff as fallback. Use when asked to review a PR, a feature branch, or pre-push changes. Read-only on code; writes a review report. Issues APPROVE or…

not rated 2 1mo ago A 82 tokens original MIT

test-quality-auditor

254

choiyounggi/loop-orchestrator

Agent

Part of loop-orchestrator

Read-only verifier that audits one task's diff and tests for quality. Invoked between self-review and done so the session that wrote the code does not grade its own tests (self-grading guard). Returns a fixed VERDICT and REASONS.

not rated 2 1mo ago A 56 tokens original MIT

implementer

255

IgnacioMarin402/spec-flow-plugin

Agent

Part of spec-flow

Implements the current milestone (Sonnet) following plan.md and the repo's own conventions. Writes code and tests. Escalates hard/complex design decisions to the architect instead of guessing. Never runs lint or tests itself — an external gate does that.

not rated 2 3d ago A 55 tokens original MIT

planner

256

IgnacioMarin402/spec-flow-plugin

Agent

Part of spec-flow

Senior planner (Opus). Turns an approved spec into a milestone-by-milestone implementation plan (M1..Mn). Also acts as the escalation consultant for the reviewer and the re-planner when the lint/test gate fails.

not rated 2 3d ago A 48 tokens original MIT

spec-writer

257

IgnacioMarin402/spec-flow-plugin

Agent

Part of spec-flow

Owns the spec artifacts (Sonnet). MODE=SPEC turns a free-text requirement into a spec with user stories and requirement deltas. MODE=TRIAGE classifies a defect by what it does to specs/. MODE=FOLD verifies a shipped change landed in the capability specs under specs/, stamps its status and archives it. Asks the human…

not rated 2 3d ago A 88 tokens original MIT

ArdurAI/ardur-agent

Agent

This document defines how Ardur Agent work is coordinated across LLM agents, human operators, and parallel coding sessions.

not rated 2 6d ago A 0 tokens original Apache-2.0

portfolio-manager

264

haiiibin/claude-multi-agent-investing

Agent Claude Code

Final synthesis judge -- receives all analyst + persona + technical + risk agent outputs, produces a decisive 5-level conviction rating (Buy/Overweight/Hold/Underweight/Sell) with account-routing, position-size guidance, and tax-aware action plan. Adapted from TauricResearch/TradingAgents portfoliomanager. Use after…

not rated 2 12d ago A 76 tokens original MIT

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: