tuicr skill
The tuicr skill links an agent to terminal code review: it reads the human's comments and opens the review pane.
- 3,601 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The agent can add comments to a review session, and `:submit` publishes a review to a hosting system as the user.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/tuicrDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The agent can add comments to a review session, and `:submit` publishes a review to a hosting system as the user.
This is third-party code. Review the repository files before installing.
What it does
tuicr is a terminal diff reviewer with vim keys, line-level comments and review submission to GitHub, GitLab, Gitea, Bitbucket, Azure DevOps or Gerrit. The `skills/tuicr` skill teaches an agent to work with saved sessions through `tuicr review`: find an open session, read the user's comments and respond to them. There are two modes. In the first, the human reviews the agent's patch while the agent only fetches comments and never writes its own. In the second, the agent reviews a patch and adds findings under an explicit agent username. With no session open, the skill launches tuicr in a cmux, Rex, Zellij or Herdr pane through wrapper scripts.
Who it is for. Developers who review agent-written code in the terminal and want comments to flow back to the agent without manual copying.
Good fit when
- An agent finished a change and the patch is read line by line in the terminal.
- After review, the agent fetches comments with `tuicr review comments`.
- The team reviews in the terminal and sends notes to a GitLab MR or Gerrit.
- An agent opens the review pane in Zellij or cmux on its own.
Not a fit when
- Review happens in the GitHub web interface with no terminal.
- You want automated checking with no human in the loop.
- No tuicr sessions exist and you do not want to install the program.
Example request
Open my changes in tuicr for review, and when I am done, fetch the comments and fix the code.Limitations
The skill works only alongside the installed tuicr program and does not display a patch on its own. The author treats an active session as a convenience signal rather than a stable protocol, so with several sessions the agent asks which one to use. cmux, Rex, Zellij and Herdr panes are covered by wrapper scripts; elsewhere the human starts the session. The last commit is dated 5 November 2026 and the tool changes quickly. The program installs separately and the wrapper scripts assume Unix shells.
How to disable. Remove the `skills/tuicr` folder from the agent's skills directory. Uninstall the program the way you installed it and clear saved sessions separately.
FAQ
How do the two skill modes differ?
In human mode the agent only reads comments through `tuicr review comments`. In agent mode it reviews the patch and adds findings with `tuicr review add` and an explicit `--username`.
How do I find the right session?
Run `tuicr review list --repo /path/to/repo` or `tuicr review list --all`. Each row carries a kind (`local` or `pr`) and a slug for the `--session` option.
Where does a finished review go?
The `:submit` command in tuicr sends it to GitHub, GitLab, Gitea, Bitbucket, Azure DevOps or Gerrit. You can also copy the markdown to the clipboard or print it to stdout.
Which version control systems work?
git, jj and Mercurial. tuicr reviews uncommitted changes, a commit range or a pull or merge request.
Do I write comments myself in human review mode?
Yes, the human writes comments in tuicr with the `c` key. The agent only reads them and never impersonates the user with its own notes.
Discussion
No questions or reviews yet.
Related
SkillsCritty opens a terminal interface where a person comments on a diff, answers questions and approves agent actions.
A terminal diff viewer: all files in one stream, and an agent with the hunk-review skill walks through the changes and leaves line comments
A TUI that shows what Claude Code, Codex, Gemini CLI and opencode do in a project: the open file, last-commit diff, live commands
Terminal diff review tool: comments live locally and a human in the TUI and an agent via the CLI share one session
A CLI opens diffs and documents in a terminal so a person can annotate lines and pass feedback to a coding agent
A terminal UI that shows everything differing from the base branch as a pull request and sends selected code to an AI agent.
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| tuicr (this card) | no | 3,601 | |
| code-review-skill | no | 2,256 | gives an agent an automated review checklist; tuicr keeps the human as reviewer in the terminal and passes their comments to the agent |
| open-code-review | no | 50,109 | builds review around model-driven checks; tuicr builds it around a human session and submission to Git hosts |
README in short
The program shows one continuous diff stream, comments on lines and ranges, reviewed marks and three ways to output a review: to a hosting system, the clipboard or stdout.