For teams
Product
Inside Glueprint
Product tour The full workflow, in 16 narrated chapters Workspaces Sessions, code, and terminals together Editor & IDE Edit, preview, and debug beside your agents Tasks & boards Plan the work and hand it to agents Channels People and agents in one conversation Assistants Memory, routines, and ongoing work Git & review Review changes, branches, and GitHub Orchestration Split one job across many agents
Get started
Desktop app Download for macOS, Windows, or Linux Mobile app Your agents on iOS and Android Always-on hosts Set up a VPS or Mac mini with the CLI Supported agents Find the agents you already use
Explore the four steps of AI adoption
Security Docs Pricing FAQ Download free Sign in
Documentation
Changelog
D-2503Settings

Agents

Which coding agents Glueprint has detected, and how to add or remove them.

Available on
Available onDesktop
Reviewed
October 15, 2026
On this page

The Agents section lists every coding agent Glueprint has detected on your machine. You point Glueprint at the agents; the agents do the actual coding work.

Supported agents

  • Claude Code — from Anthropic. Conversation, full tool use, session resume.
  • Codex — from OpenAI. Conversation, tool use, mid-turn steering.
  • Gemini CLI — from Google. Conversation, tool use.

Each agent is a CLI you install yourself. Glueprint doesn’t bundle them.

Auto-detection

When Glueprint launches, it scans your PATH for known agent binaries. Detected agents show up in the Agents list with their version and the path Glueprint will use.

If an agent isn’t detected:

  1. Install the agent’s CLI per its vendor’s instructions.
  2. Make sure it’s on PATH for the shell Glueprint launches with (usually a login shell on macOS and Linux).
  3. Open the agent’s card in Settings > Agents and click Detect to re-scan that agent, or relaunch the app to re-scan all agents.

Custom paths

If an agent is installed at a non-standard location, click Edit on its row and paste the full path. Glueprint will use that path instead of the auto-detected one.

Default agent

The first agent in the list is Glueprint’s default for new sessions. Drag to reorder. The reorder applies everywhere — the Sessions composer, the assistant create form, the routine setup.

What Glueprint does with each agent

When you start a session, Glueprint:

  1. Spawns the agent as a child process inside the working directory.
  2. Pipes your initial prompt and any context through the agent’s standard input.
  3. Reads everything the agent produces from its standard output.
  4. Renders messages, tool calls, and file edits in the session view.

Each agent has slightly different capabilities; Glueprint accommodates the differences so you don’t have to think about them. The agents’ configuration (API keys, model selection, rate limits) is managed where the agent itself wants — usually in its own config file or environment variables.

Signing agents in

Each agent CLI keeps its own login on the host it runs on. When one is signed out, its card shows Needs login; click Log in and Glueprint runs the agent’s own sign-in for you, or choose Use API key. This works for agents on other hosts too: see Sign agents in on another host.

  • Claude Code — signs in with your Claude account, or uses ANTHROPIC_API_KEY.
  • Codex — signs in with your ChatGPT account (device code), or an OpenAI API key.
  • Gemini CLI — signs in with Google, or uses GEMINI_API_KEY.

Your usage cost is whatever the agent vendor charges; Glueprint doesn’t markup or relay.

Multi-machine

Agents are per-host. If you install Glueprint on a second machine, install the agents on that machine too. Settings > Agents lists the agents on the machine you’re on; to see or sign in the agents on another host, open that host’s settings (the gear next to it in the host picker) and choose Agents.