Supported Agents

Gump orchestrates coding agent CLIs. Each agent runs in headless mode inside the worktree. Gump prepares the context, launches the process, streams its output, and collects the result.

Agent table

Agent	CLI command	Context file	Resume	Tokens	Native cost
Claude Code	`claude -p`	CLAUDE.md	`--resume <uuid>`	Per message	Yes
Codex	`codex exec`	AGENTS.md	`resume <thread-id>`	Per turn	No
Gemini CLI	`gemini -p`	GEMINI.md	`--resume`	Aggregated	No
Qwen CLI	`qwen -p`	QWEN.md	`--resume <uuid>`	Per message	No
OpenCode	`opencode run`	AGENTS.md	`--session <ses_id>`	Per step_finish	No
Cursor	`cursor-agent -p`	`.cursor/rules/*.mdc`	`--resume [chatId]`	Per message	No

Using agents in workflows

Reference agents by name in your workflow YAML:

- name: impl
  type: code
  run:
    agent: claude-sonnet

Check which agents are available:

gump models

Compat mode

If an agent provider changes its output format or breaks its CLI, Gump falls back to compat mode: raw stdout (not parsed for metrics), git diff as the source of truth, fresh sessions only. The workflow still works — you just lose granular metrics until the adapter is updated.

Adapter doctrine

“Gump executes agents through their native CLI surface. Each adapter encapsulates provider-specific behavior. Gump never reimplements agent reasoning or tool routing.”

Connector roadmap

Wave	Agents	Status
Wave 1	Claude Code, Codex, Gemini, Qwen, OpenCode	Available
Wave 1.5	Cursor CLI	Available
Wave 2	Aider, Copilot CLI, Amp, Kimi CLI	Planned
Wave 3	Trae CLI, Neovate, Fetchcoder	Evaluating

Overview

Getting Started

Writing Workflows

Orchestration

Execution & Data

Agents

Reference

Supported Agents

Supported Agents

Agent table

Using agents in workflows

Compat mode

Adapter doctrine

Connector roadmap

​Supported Agents

​Agent table

​Using agents in workflows

​Compat mode

​Adapter doctrine

​Connector roadmap

Supported Agents

Agent table

Using agents in workflows

Compat mode

Adapter doctrine

Connector roadmap