Agents | Aigon

Aigon supports four AI coding agents. Each gets its own slash command prefix, settings, and context delivery mechanism.

Supported agents

Code	Agent	Slash prefix	CLI command
`cc`	Claude Code	`/aigon:`	`claude`
`gg`	Gemini CLI	`/aigon:`	`gemini`
`cx`	Codex CLI	`/prompts:aigon-`	`codex`
`cu`	Cursor	`/aigon-`	`agent`

Installing agents


aigon install-agent cc gg          # Install Claude + Gemini
aigon install-agent cx cu          # Add Codex + Cursor later

Install only the agents you have access to. You can add more at any time.

What install-agent writes

aigon install-agent writes only aigon-owned files. It never touches user-owned root files like CLAUDE.md.

Per-agent files

Agent	Slash commands	Settings	Context delivery
cc	`.claude/commands/aigon/*.md`	`.claude/settings.json`	`.claude/skills/aigon/SKILL.md` + SessionStart hook
gg	`.gemini/commands/aigon/*.toml`	`.gemini/settings.json`	SessionStart hook
cx	`~/.codex/prompts/aigon-*.md` (global)	`.codex/config.toml`	`.codex/prompt.md`
cu	`.cursor/commands/aigon-*.md`	`.cursor/cli.json`, `.cursor/hooks.json`	`.cursor/rules/aigon.mdc`

Shared files (all agents)

File	Created	Updated on re-install
`AGENTS.md`	Scaffolded on first install only	Never (your file)
`docs/agents/{agent}.md`	Yes	Yes (marker blocks only)
`docs/development_workflow.md`	Yes	Yes (full overwrite)

How context reaches each agent

Instead of injecting into user-owned files, Aigon delivers context through each agent’s native extension mechanism:

Claude Code / Gemini CLI: A SessionStart hook runs aigon project-context, which prints doc pointers to stdout. The agent ingests this as conversation context.
Cursor: .cursor/rules/aigon.mdc with alwaysApply: true
Codex: .codex/prompt.md with marker blocks

Updating


aigon update

Re-runs install-agent for all detected agents. Updates command templates, hooks, and aigon-owned doc files. Never touches CLAUDE.md or AGENTS.md.

Auto-update: The aigon check-version SessionStart hook detects version mismatches and runs aigon update automatically.

Project-specific instructions

Add shared project rules directly in your root instruction files:

AGENTS.md — Read by Gemini, Codex, and Cursor
CLAUDE.md — Claude Code’s native project instructions
docs/agents/{id}.md — Agent-specific operational notes (aigon-managed)

Permission modes

Default “yolo mode” flags auto-approve commands:

Agent	Default flag	Effect
`cc`	`--permission-mode acceptEdits`	Auto-edits, prompts for risky Bash
`cu`	`--force`	Auto-approves commands
`gg`	`--yolo`	Auto-approves all
`cx`	(empty)	Interactive by default (`--full-auto` only in autonomous mode)

Override with:


aigon config set --global agents.cc.implementFlag ""

Using agents in Fleet mode

In Fleet mode, each agent gets an isolated worktree. Mismatch protection prevents launching the wrong agent in a worktree:


feature-108-cc-dark-mode/  → only cc can run here
feature-108-gg-dark-mode/  → only gg can run here
feature-108-cx-dark-mode/  → only cx can run here

Terminal integration

How agents launch depends on your terminal config:

Terminal	Agent auto-launch	Persistent sessions
Warp	Yes (split panes)	No
tmux	Yes (named sessions)	Yes
VS Code / Cursor	Manual	No
Terminal.app	Yes (new window)	No

Configure with:


aigon config set --global terminal tmux
aigon config set --global tmuxApp iterm2