Stax Dev
cesarferreira/stax
Development harness for the stax Rust CLI project. An agent skill from cesarferreira/stax.
Drive and debug the real keifu TUI autonomously via its remote-control debug server (--debug-listen) — launch headlessly, inject keys/mouse, dump the rendered screen as text, and inspect app state.
Install Debug Tui
$ npx skills add trasta298/keifu --skill debug-tui -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install trasta298/keifu debug-tui --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/trasta298/keifu.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/debug-tui .claude/skills/debug-tui && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "debug-tui" agent skill from https://github.com/trasta298/keifu/tree/main/.agents/skills/debug-tui into .claude/skills/debug-tui/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "debug-tui", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/trasta298/keifu/tree/main/.agents/skills/debug-tuiType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add trasta298/keifu --skill debug-tui -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install trasta298/keifu debug-tui --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/trasta298/keifu.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/debug-tui .agents/skills/debug-tui && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "debug-tui" agent skill from https://github.com/trasta298/keifu/tree/main/.agents/skills/debug-tui into .agents/skills/debug-tui/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "debug-tui", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add trasta298/keifu --skill debug-tui -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install trasta298/keifu debug-tui --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/trasta298/keifu.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/debug-tui .cursor/skills/debug-tui && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "debug-tui" agent skill from https://github.com/trasta298/keifu/tree/main/.agents/skills/debug-tui into .cursor/skills/debug-tui/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "debug-tui", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/trasta298/keifu.git --path .agents/skills/debug-tui--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add trasta298/keifu --skill debug-tui -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install trasta298/keifu debug-tui --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/trasta298/keifu.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/debug-tui .gemini/skills/debug-tui && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "debug-tui" agent skill from https://github.com/trasta298/keifu/tree/main/.agents/skills/debug-tui into .gemini/skills/debug-tui/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "debug-tui", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install trasta298/keifu debug-tuiInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add trasta298/keifu --skill debug-tui -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/trasta298/keifu.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/debug-tui .github/skills/debug-tui && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "debug-tui" agent skill from https://github.com/trasta298/keifu/tree/main/.agents/skills/debug-tui into .github/skills/debug-tui/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "debug-tui", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add trasta298/keifu --skill debug-tui -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install trasta298/keifu debug-tui --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/trasta298/keifu.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/debug-tui .opencode/skills/debug-tui && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "debug-tui" agent skill from https://github.com/trasta298/keifu/tree/main/.agents/skills/debug-tui into .opencode/skills/debug-tui/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "debug-tui", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
debug-tui- GitHub stars
- 811
- Token cost
- ~1.3k tokens
- SKILL.md length
- 600 words
- Files
- 1
- Skills in repo
- 2
- Repo updated
- First seen
- Licence
- MIT
At a glance
Drive and debug the real keifu TUI autonomously via its remote-control debug server (--debug-listen) — launch headlessly, inject keys/mouse, dump the rendered screen as text, and inspect app state.
- Works in 4 steps: dump and confirm the precondition is on… → keys / mouse to act. → state + dump, then assert: grep the dump… → …
- Tasks that involve Debugging
- SKILL.md covers Launch, Protocol, Gotchas that will waste your… and Verification loop
- Calls cargo, git and jq
What it does
Debug Tui is an agent skill from trasta298/keifu. Drive and debug the real keifu TUI autonomously via its remote-control debug server (--debug-listen) — launch headlessly, inject keys/mouse, dump the rendered screen as text, and inspect app state. Use this whenever a change affects TUI behavior, rendering, keybindings, mouse handling, focus, scrolling, or async loading states, when reproducing a user-reported UI bug, or when you need to confirm "does it actually work on screen" — cargo test alone cannot verify what the user sees. Reproduce the issue through this…
Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Development, covering Debugging. It works with Git and Rust. The repository describes itself as: Git genealogy, untangled. A TUI for navigating commit graphs with color and clarity. The licence is MIT.
When your agent uses it
- Tasks that involve Debugging
Example prompts
- “does it actually work on screen”
- “/debug-tui”
Workflow steps
4 steps, taken from the first numbered list in SKILL.md.
- dump and confirm the precondition is on screen (e.g. the row you'll click).
- keys / mouse to act.
- state + dump, then assert: grep the dump for expected text, compare
- Quit, and read /tmp/keifu.log for the tracing trail
What it can do on your machine
Read from SKILL.md and the folder at commit 4d9f033. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
cargogitjq
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Debug Tui loads about 1.3k tokens when it runs. Until then it costs about 143 tokens; SKILL.md has 600 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from trasta298/keifu at commit 4d9f033, republished under its MIT licence (© trasta298). 600 words, ~1,312 tokens.
Download SKILL.mdSave it as
.claude/skills/debug-tui/SKILL.md (or your agent's skills folder).- name
- debug-tui
- description
- Drive and debug the real keifu TUI autonomously via its remote-control debug server (--debug-listen) — launch headlessly, inject keys/mouse, dump the rendered screen as text, and inspect app state. Use this whenever a change affects TUI behavior, rendering, keybindings, mouse handling, focus, scrolling, or async loading states, when reproducing a user-reported UI bug, or when you need to confirm "does it actually work on screen" — cargo test alone cannot verify what the user sees. Reproduce the issue through this workflow before fixing, and re-verify after.
Debugging the keifu TUI headlessly
keifu has a built-in remote-control server. You can run the real app without a human at the terminal: send key/mouse input through the same code paths as real input, dump the rendered screen as plain text, and read the app state as JSON. The reliable workflow is always: drive → dump → assert, both to reproduce a bug and to prove the fix.
Launch
cargo build
PORT=7167 # pick a fresh port per run to avoid stale instances
timeout 120 script -qec "./target/debug/keifu --debug-listen 127.0.0.1:$PORT --log-file /tmp/keifu.log" /dev/null >/dev/null 2>&1 &
sleep 2scriptallocates a PTY; keifu cannot enable raw mode without one. Thetimeoutwrapper guarantees stray instances die even if you forget to quit.- The
scriptPTY reports size 0x0, so the main loop skips real rendering. Consequence: pane layout for mouse hit-testing is only recorded when a render happens — always send adumpwith explicit width/height before anymousecommand, and give mouse coordinates in that dump's space. - keifu operates on the repository of its working directory. To exercise
staging/commit/push, launch it inside a throwaway repo (
mktemp -d+git init), never the real working repo.
Protocol
Newline-delimited JSON over TCP; each request line gets one JSON response line.
printf '%s\n' '{"cmd":"state"}' | nc -q1 127.0.0.1 $PORT| Request | Effect |
|---|---|
{"cmd":"keys","keys":"j j <enter>"} | Inject key input (normal keybinding layer) |
{"cmd":"mouse","kind":"click","x":5,"y":3} | Click / scroll_up / scroll_down at 0-based cell |
{"cmd":"dump","width":110,"height":30} | Render current state to plain text at that size |
{"cmd":"state"} | Mode, focused pane, selection, HEAD, async status |
For "feels slow" reports, use the log: ops over 10ms are written live as
slow operation, and quitting writes a per-op perf summary (count/avg/max).
Reproduce → quit → grep the log file.
Every response is one JSON line. dump returns the screen as an escaped
string in the screen field — pipe through jq -r .screen to read it. For
single requests prefer nc -q1 (closes after the response); only multi-line
batches need plain nc under timeout.
Key token syntax: whitespace-separated; single chars as-is (uppercase implies
Shift); special keys <enter> <esc> <tab> <backtab> <space> <up> <down> <left> <right> <home> <end> <pgup> <pgdn> <backspace> <c-x> (Ctrl+x). To type a word
in an input dialog, space-separate the letters: c f i x <space> b u g <enter>.
Full protocol details: docs/debugging.md. Implementation: src/debug_server.rs.
Show full SKILL.md (272 more words)Show less
Gotchas that will waste your time
Double-click = two clicks on the same cell within 400 ms. Separate
ncinvocations are too slow — send both clicks (plus the leadingdumpthat records the layout) in ONE connection:bash printf '%s\n%s\n%s\n%s\n' \ '{"cmd":"dump","width":110,"height":30}' \ '{"cmd":"mouse","kind":"click","x":60,"y":24}' \ '{"cmd":"mouse","kind":"click","x":60,"y":24}' \ '{"cmd":"state"}' | timeout 4 nc 127.0.0.1 $PORTCommands are processed after the event-poll tick, so responses can lag up to ~200 ms; wrap
ncintimeoutand don't interpret slowness as a hang.A held-open
ncmay exit non-zero viatimeouteven after delivering the response — check the output, not the exit code.Injected input bypasses the terminal's input layer.
keys/mousecommands go straight into the app, so they cannot verify anything that depends on terminal modes — e.g. mouse tracking escape sequences (?1000/?1002/?1003) set insrc/tui.rs. Changes there need a human in a real terminal.qonly quits from the graph pane. If another pane is focused or a popup is open (e.g. after a mouse click),q/<esc>first returns focus/closes the popup and the app keeps running. Send{"cmd":"keys","keys":"q q"}and confirm exit: a follow-upncconnection must be refused. (pgrep -af keifumatches your own shell's command line — don't trust it.)
Verification loop
dumpand confirm the precondition is on screen (e.g. the row you'll click).keys/mouseto act.state+dump, then assert: grep the dump for expected text, compareselected_index/mode/focused_panein the state JSON.- Quit, and read
/tmp/keifu.logfor the tracing trail (KEIFU_LOG=tracefor more detail; useful for async diff-load issues).
A fix is not verified until step 3 shows the corrected behavior on a dump that previously showed the bug.
© trasta298, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in .agents/skills/debug-tui of trasta298/keifu.
Open the folder on GitHubat commit 4d9f033
Compare with similar skills
Debug Tui next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Debug Tui this skilltrasta298/keifu | 811 | — | ~1.3k | Automated safety check: Pass | MIT | |
| Stax Devcesarferreira/stax | 128 | — | ~987 | Automated safety check: Pass | MIT | |
| Debuggnomeria/usbtree | 688 | — | ~715 | Automated safety check: Pass | MIT | |
| OpenLogi macOS Permissions TriageAprilNEA/OpenLogi | 23k | — | ~2.5k | Automated safety check: Notes | Apache-2.0 | |
| Worktrunk Tend CI Guidancemax-sixty/worktrunk | 8.9k | — | ~6.4k | Automated safety check: Pass | Custom licence | |
| Debugging and Error Recoveryaddyosmani/agent-skills | 102k | 1 repos | ~2.6k | Automated safety check: Pass | MIT |
Similar skills
Debug
gnomeria/usbtree
Systematic root-cause debugging — reproduce, isolate, fix at the source, prove the fix.
OpenLogi macOS Permissions Triage
AprilNEA/OpenLogi
Decides whether an OpenLogi device problem on macOS is a privacy-permission (TCC) problem, using agent log lines, and says which identity needs which grant.
Worktrunk Tend CI Guidance
max-sixty/worktrunk
Adds Worktrunk-specific rules to the tend CI workflows: Codecov polling, Rust test commands, labels and review criteria for pull requests handled in CI.
Debugging and Error Recovery
addyosmani/agent-skills
Applies a stop-the-line rule and a step-by-step triage when tests fail, builds break or something stops working, aiming at the root cause instead of guesses.
Worktrunk Release Workflow
max-sixty/worktrunk
Walks a maintainer through cutting a Worktrunk release: sync the release branch, pass two test gates, review the changes, then publish.
More from trasta298/keifu
Release
trasta298/keifu
Create a new release with tag, GitHub release, and homebrew-tap update
Related
Categories
Questions about Debug Tui
What does Debug Tui do?
Drive and debug the real keifu TUI autonomously via its remote-control debug server (--debug-listen) — launch headlessly, inject keys/mouse, dump the rendered screen as text, and inspect app state. Debug Tui is an agent skill from trasta298/keifu. Drive and debug the real keifu TUI autonomously via its remote-control debug server (--debug-listen) — launch headlessly, inject keys/mouse, dump the rendered screen as text, and inspect app state.
When should I use Debug Tui?
Debug Tui fits situations like: tasks that involve Debugging.
How do I install Debug Tui in Claude Code?
Run `npx skills add trasta298/keifu --skill debug-tui -a claude-code`. Or copy the skill folder (.agents/skills/debug-tui in trasta298/keifu) into .claude/skills/debug-tui in your project. Claude Code loads it when a task matches its description.
How do I install Debug Tui in Codex?
Run `npx skills add trasta298/keifu --skill debug-tui -a codex`. Or copy the skill folder (.agents/skills/debug-tui in trasta298/keifu) into .agents/skills/debug-tui in your project. Codex loads it when a task matches its description.
Can I use Debug Tui in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add trasta298/keifu --skill debug-tui -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/debug-tui, .gemini/skills/debug-tui, .github/skills/debug-tui and .opencode/skills/debug-tui in your project.
What does Debug Tui need to run?
Going by SKILL.md and its folder, Debug Tui needs the command-line tools its instructions call (cargo, git and jq).
Does Debug Tui access the network?
SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Debug Tui safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Debug Tui use?
Debug Tui is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Debug Tui use?
About 1.3k tokens (SKILL.md is roughly 5.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Debug Tui?
Skills that share tags, products or a category with Debug Tui: Stax Dev (cesarferreira/stax, 128 stars), Debug (gnomeria/usbtree, 688 stars), OpenLogi macOS Permissions Triage (AprilNEA/OpenLogi, 23k stars) and Worktrunk Tend CI Guidance (max-sixty/worktrunk, 8.9k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Debug Tui?
trasta298 (a GitHub user) maintains it in trasta298/keifu, which has 811 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on July 30, 2026.
Source: trasta298/keifu on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.