I recently transitioned to using the new Antigravity CLI (agy), which has officially replaced the older Gemini CLI as the main terminal interface for agentic coding workflows. Here are some of my initial thoughts on my experience using it.

The Good Bits

• Performance: It’s generally faster and much more responsive than the old Gemini CLI. Command execution and responses feel near-instant.
• Context breakdown: The /context command displays a clean breakdown of loaded files and symbols, making it easy to track what is currently in the agent’s context window.
• Conversation resets: The /clear (or /new) command resets the conversation history without requiring you to quit and restart the CLI session.
• Sub-agents: Spawning sub-agents to handle background research or run A/B tests (like comparing two different implementations) is extremely easy and doesn’t interrupt the main thread.

Context Bloat & other issues

• In sequential coding workflows, context windows bloat rapidly as tool execution outputs and file reads are appended to history, causing latency and token waste. A big multi-file refactoring session can run out of tokens quickly because of the lack of a compression command. The older Gemini CLI resolved this with the /compress command, which replaced the chat context with a high-level summary of the conversation. Similar developer interfaces (like OpenCode with /compact) implement a matching pattern to keep reasoning loops tight.

• Another problem is that agy can often mess up the application of moderate-to-large diffs; it sometimes removes more lines than expected, or adds extra lines that result in syntax errors.

• The request-review mode for tool permission is not “smart enough”. Many times I have commands that I don’t want to apply globally, but at least in a session I might want to have it allowed. Antigravity often suggest 2 options, one is to allow the exact command, and another to add the exact command to settings, but it doesn’t have the option to allow the command general pattern, such as “cp *” in this particular session and that can become annoying as I have to approve the command again and again for slightly different parameters

• Another issue related to the request-review mode is that the list of choice can be messed up if the choice is python3 run -c ..... Here, the LLM is using a python script as a string literal on the command line, so the choice is a huge block of text and therefore I can only see the last 2 choice and I can’t even scroll up to see what the first choices are

Workaround for context bloat

Currently, agy supports /rewind (to undo the session to a previous checkpoint) but has no native /compact or /compress command. The current workaround requires manually managing state:

1. Keeping a tracking file like todo.md.
2. Having the agent update it after each task.
3. Clearing the session history via /new.
4. Reading todo.md in the new session to resume.

This workaround works, but having to manually cycle through session restarts feels unnecessarily clunky. Context hygiene should be a native feature of the harness.