agent-loop
agent-loop is an MCP server, listed under model apis. It reaches anthropic, brave, confluence, jira and openai. It needs ANTHROPIC_API_KEY, BRAVE_API_KEY, CONFLUENCE_API_TOKEN, JIRA_API_TOKEN and OPENAI_API_KEY to authenticate. Its repository has 28 GitHub stars and is licensed under NOASSERTION; we last read it on 2026-10-22.
Agent-loop connects to Model Context Protocol (MCP) and provides an AI Agent with optional human oversight, featuring a customizable command-line interface.
Install
MCP servers are packaged differently from each other -- some run via npx, some as a Docker image, some as a plain script you clone and run. agent-loop’s own README did not state one plainly enough for this page to quote it rather than guess, so it will not guess. The repository below states it.
What it reaches, and what it needs
Maintenance
An AI Agent with optional Human-in-the-Loop Safety, Model Context Protocol (MCP) integration, and beautiful, themeable CLI output
What agent-loop can do
8 tools documented in its README. An AI client calls these by name.
bash- No description in the README.
cli_plot- No description in the README.
list_dir- No description in the README.
codebase_search- No description in the README.
file_search- No description in the README.
grep_search- No description in the README.
project_inspector- No description in the README.
aws_cli- No description in the README.
Read from the README; names are kept only if the README contains them.
agent-loop compared with similar servers
| Server | Reaches | Needs | Stars | Last update | Licence |
|---|---|---|---|---|---|
| agent-loop (this page) | anthropic, brave, confluence, jira, openai | ANTHROPIC_API_KEY, BRAVE_API_KEY, CONFLUENCE_API_TOKEN, JIRA_API_TOKEN, OPENAI_API_KEY | 28 | 2026-01-12 | NOASSERTION |
| just-prompt | anthropic, deepseek, gemini, groq, openai(2 shared) | ANTHROPIC_API_KEY, DEEPSEEK_API_KEY, GEMINI_API_KEY, GROQ_API_KEY, OPENAI_API_KEY | 739 | 2025-08-10 | — |
| flama | anthropic, openai(2 shared) | None listed | 304 | 2026-10-13 | Apache-2.0 |
| schemagate | anthropic, entra, gemini, mysql, openai, postgres, schemagate(2 shared) | ANTHROPIC_API_KEY, ENTRA_CLIENT_SECRET, GEMINI_API_KEY, OPENAI_API_KEY, SCHEMAGATE_AUDIT_SIGNING_KEY | 105 | 2026-10-22 | Apache-2.0 |
| agent-fleet-o | anthropic, google, openai, typesafe(2 shared) | ANTHROPIC_API_KEY, DB_PASSWORD, GOOGLE_AI_API_KEY, OPENAI_API_KEY, OP_SERVICE_ACCOUNT_TOKEN, TYPESAFE_API_KEY | 76 | 2026-10-20 | AGPL-3.0 |
| Promptheus | anthropic, google, openai(2 shared) | ANTHROPIC_API_KEY, GOOGLE_API_KEY, OPENAI_API_KEY | 17 | 2026-10-13 | MIT |
Ranked by services in common with agent-loop, then by stars. Figures are from each repository when we last read it.
Common questions about agent-loop
- What does agent-loop do?
- Agent-loop connects to Model Context Protocol (MCP) and provides an AI Agent with optional human oversight, featuring a customizable command-line interface.
- How do I install agent-loop?
- Its README did not state an install command plainly enough for us to quote one, so we do not guess. The repository has the instructions: https://github.com/AlessandroAnnini/agent-loop
- What does agent-loop connect to, and what does it need?
- It reaches anthropic, brave, confluence, jira and openai. The credentials we found in its source are ANTHROPIC_API_KEY, BRAVE_API_KEY, CONFLUENCE_API_TOKEN, JIRA_API_TOKEN and OPENAI_API_KEY.
- Is agent-loop official and is it maintained?
- We found nothing naming it as a vendor’s official server. Its repository has 28 GitHub stars and its state in our directory is listed. Its last commit we recorded was on 2026-01-12.
- What licence does agent-loop use?
- The repository declares the NOASSERTION licence.
More MCP servers for model apis
- wigolo5,498 stars
- KiCAD-MCP-Server2,905 stars
- modelcontextprotocol2,589 stars
- paperbanana2,412 stars
- gemini-mcp-tool2,294 stars
- telegram-mcp1,879 stars