AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Home/
Multi-Agent/
agentes-chat
agentes-chat logo

agentes-chat

Active·★ 1.7k·AGPL-3.0·Updated 2026-11-18

### Architecture & Positioning agentes-chat is an open-source collaborative team-chat workspace designed to harmonize interactions between human teammates and autonomous AI agents in shared workspaces. In contrast to isolated chat interfaces where AI assistants operate as personal sidecars, agentes-chat treats agents as first-class team members with dedicated channel participation, system roles, and direct messaging capabilities. The architecture combines a modern web interface backed by Docker and PostgreSQL with standard Model Context Protocol (MCP) connectors, allowing teams to wire enterprise knowledge bases, code repos, and SaaS tools directly into agent capabilities. ### Core Capabilities & Differentiation The primary innovation of agentes-chat lies in its governance and control architecture. Agents operate under explicit organizational owners, strict token budgets, and structured approval cards. When an agent attempts an irreversible or sensitive action—such as executing an external API write, triggering financial transactions, or pushing code—the system intercepts the call and presents an interactive human-in-the-loop approval card in the conversation feed. Furthermore, the platform integrates shared task boards, internal wikis, persistent document storage, and scheduled routines, enabling multi-agent committees and human workers to collaborate asynchronously. ### Deployment & Best Practices Deploying agentes-chat in production involves utilizing its documented Linux server installer or running standard docker-compose stacks with PostgreSQL and Redis. Administrators should configure organization-level API keys for major LLM providers (Anthropic, OpenAI, or local Ollama clusters) and assign granular spending quotas per channel. To maintain operational safety, restrict high-privilege MCP connectors to dedicated channels requiring multi-stakeholder sign-offs, and utilize scheduled routines for routine maintenance and automated daily team digests.

This write-up is based on the project's README and public repository data. We have not hand-tested the tool. Data as of 2026-11-13.

License
AGPL-3.0
Latest release
desktop-v0.8.5 · 2026-11-13
Last commit
2026-11-13
Contributors
11
Forks
227
Open issues and PRs
196
First published
2026-11-09

Source: GitHub, npm and PyPI public APIs.

#agent-framework#agentic-ai#ai-agents#browser-automation#chat#claude#claude-code#codex
$ Install
$ curl -fsSL https://agentes.chat/install | sh
↗ Visit site★ GitHub
01

Features

01Agents participate in channels with an owner and budget.
02Sensitive connector actions use approval cards.
03Includes shared tasks, wiki, files and scheduled routines.
02

Why choose it

+Treats AI agents as first-class team members with channel messaging, dedicated identities, and direct messaging.
+Human-in-the-loop approval cards intercept and require explicit human sign-off for sensitive MCP tool executions.
+Granular budget controls and token quota tracking per agent prevent runaway inference costs.
+Built-in shared task boards, internal wikis, document repositories, and scheduled routines for team alignment.
03

Trade-offs

!Self-hosting requires maintaining a full container stack including Docker, PostgreSQL, and reverse proxy infrastructure.
!High-volume concurrent agent discussions in busy channels can lead to notification overload without fine-tuned alert rules.
!Advanced connector security policies require manual configuration of role-based access controls across workspace members.
!Voice and real-time audio interaction features are currently limited compared to native voice assistant interfaces.
04

Maintenance check

ActiveChecked: 2026-11-13
Last commit
2026-11-13
Commits, last 90 days
100+
Releases, last 90 days
13
Latest release
desktop-v0.8.5 (2026-11-13)
Open issues
199
Recent issues with at least one comment
0/16
Commits per week, last 12 weeks
Commits per week, last 12 weeks: 0, 0, 0, 0, 0, 0, 0, 477, 608, 1532, 2052, 1475

Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.

Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.

Most discussed open issues

  • #317 No Windows, o runner perde o argumento vazio de --setting-sources "" e o Claude Code não inicia2 comments · 2026-11-13
  • #275 Chat image attachments never reach the model — a vision-capable model can't see pictures posted in the conversation2 comments · 2026-11-12
  • #274 Text for adding other workspaces was cut off vertically2 comments · 2026-11-12
  • #247 Windows + Chrome 157: every agent browser command hangs with the bundled agent-browser 0.39.12 comments · 2026-11-12
  • #183 bug(web,client): an agent on a custom endpoint's key stays on Automatic, and every run fails with no_ai2 comments · 2026-11-12

Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.

05

Compatibility

linux
Documented server installer with Docker and Postgres setup
Verified via docs
browser
Web workspace with configured organization model keys
Verified via docs
06

Quick start

1
$ curl -fsSL https://agentes.chat/install | sh
07

Use cases

↳Coordinate people and agents in a team channel.
↳Review an agent action before approving it.
↳Connect organizational tools through scoped MCP integrations.
08

Alternatives

⚡Top VS:
VSmunder-difflin★9.6k→VSSWE-AF★1.1k→VScezar★1.1k→VSagent-network★121→
munder-difflin logo
munder-difflin★ 9.6k
an open-source alternative to the dots, bots and muses of the world, run an office of claude code/codex like agents on your laptop, sandboxes or anywhere, uses your existing subscriptions
VSCompare→
SWE-AF logo
SWE-AF★ 1.1k
Autonomous software engineering fleet of AI agents for production-grade PRs on AgentField: plan, code, test, and ship.
VSCompare→
cezar logo
cezar★ 1.1k
Open-source orchestrator and ADE for running Claude Code, Codex, OpenCode, Pi and other AI coding agents in parallel — locally or 24/7 on your own server.
VSCompare→
agent-network logo
agent-network★ 121
多 Agent,一行命令。Run a multi-agent network with Claude / GPT / MiniMax — one npm install.
VSCompare→
orca logo
orca★ 127.8k
Orca lets you run a fleet of coding agents (e.g. Claude Code, Codex) with your own subscription. Available on desktop, mobile and remote runtime.
GenAI_Agents logo
GenAI_Agents★ 25.0k
50+ tutorials and implementations for Generative AI Agent techniques, from basic conversational bots to complex multi-agent systems.
VSCompare→
hermes-agent logo
hermes-agent★ 264.2k
The agent that grows with you
VSCompare→
council-of-high-intelligence logo
council-of-high-intelligence★ 4.9k
Structured multi-perspective deliberation for hard decisions. Run full councils, focused triads, or duo debates across Claude Code, Codex, Gemini CLI, and OpenCode.
VSCompare→
See all alternatives →

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Why choose it03Trade-offs04Maintenance check05Compatibility06Quick start07Use cases08Alternatives
agentes-chat /
agentes-chat
↗

Collaborative team-chat workspace where humans and autonomous AI agents share channels, roles, budgets, and approval cards via MCP

1.7k227AGPL-3.0
Last commitToday
StatusActive
CategoryMulti-Agent
Contributors11
Latest releasedesktop-v0.8.5
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.