Senpi Agent QA Harness
code-yeongyu/senpi
Checks changes to the senpi coding agent by driving the real CLI from source in an isolated sandbox, over RPC, terminal UI, mock model and CLI smoke channels.
Runs live QA for the CodexBar app: provider usage matrix checks through its packaged CLI, config validation and menu checks, with 1Password-backed credentials handled safely.
$ npx skills add steipete/CodexBar --skill qa-test -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install steipete/CodexBar qa-test --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/steipete/CodexBar.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/qa-test .claude/skills/qa-test && 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 "qa-test" agent skill from https://github.com/steipete/CodexBar/tree/main/.agents/skills/qa-test into .claude/skills/qa-test/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qa-test", 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/steipete/CodexBar/tree/main/.agents/skills/qa-testType 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 steipete/CodexBar --skill qa-test -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install steipete/CodexBar qa-test --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/steipete/CodexBar.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/qa-test .agents/skills/qa-test && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "qa-test" agent skill from https://github.com/steipete/CodexBar/tree/main/.agents/skills/qa-test into .agents/skills/qa-test/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qa-test", 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 steipete/CodexBar --skill qa-test -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install steipete/CodexBar qa-test --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/steipete/CodexBar.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/qa-test .cursor/skills/qa-test && 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 "qa-test" agent skill from https://github.com/steipete/CodexBar/tree/main/.agents/skills/qa-test into .cursor/skills/qa-test/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qa-test", 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/steipete/CodexBar.git --path .agents/skills/qa-test--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 steipete/CodexBar --skill qa-test -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install steipete/CodexBar qa-test --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/steipete/CodexBar.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/qa-test .gemini/skills/qa-test && 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 "qa-test" agent skill from https://github.com/steipete/CodexBar/tree/main/.agents/skills/qa-test into .gemini/skills/qa-test/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qa-test", 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 steipete/CodexBar qa-testInstalls 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 steipete/CodexBar --skill qa-test -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/steipete/CodexBar.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/qa-test .github/skills/qa-test && 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 "qa-test" agent skill from https://github.com/steipete/CodexBar/tree/main/.agents/skills/qa-test into .github/skills/qa-test/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qa-test", 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 steipete/CodexBar --skill qa-test -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install steipete/CodexBar qa-test --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/steipete/CodexBar.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/qa-test .opencode/skills/qa-test && 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 "qa-test" agent skill from https://github.com/steipete/CodexBar/tree/main/.agents/skills/qa-test into .opencode/skills/qa-test/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qa-test", 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.
qa-testRuns live QA for the CodexBar app: provider usage matrix checks through its packaged CLI, config validation and menu checks, with 1Password-backed credentials handled safely.
The skill is for live provider testing, release smoke tests, menu verification and debugging reports that a provider works or fails. The agent works from the CodexBar checkout and uses the packaged `CodexBarCLI` helper rather than the app binary, which can look like it hangs as a CLI. A bundled script, `scripts/live_provider_matrix.sh`, runs the checks in several modes: the providers enabled in the app's config, the app-facing default command, every registered provider (a discovery tool that is expected to fail for providers without sessions or keys) or a list you name. A config counts as green when the enabled and default runs are clean.
Config QA validates the config file, prints its shape with API keys, secrets and cookie headers redacted, and takes a timestamped backup before any edit. Menu checks use Peekaboo after the CLI passes. For API behavior the agent browses official provider docs only, with a reference file of API specs, and may use Browser Use for logged-in dashboards. Safety rules forbid dumping environment variables or secret patterns, keep 1Password commands inside one persistent tmux session with no raw secret output, and treat browser-cookie and keychain flows as likely to trigger prompts.
Read from SKILL.md and the folder at commit dd7dca4. It shows what the files ask for, not the result of running them.
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.
Ships 1 file in scripts/ (Shell), which the agent can run.
Shell commands in SKILL.md call:
jqrgmakeFrom the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names these keys or tokens, usually read from environment variables:
OPENAI_API_KEYFrom names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
CodexBar Live QA loads about 1.2k tokens when it runs, and up to ~1.4k if it reads all its reference files. Until then it costs about 60 tokens; SKILL.md has 387 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.
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); the scripts in this folder are not scanned.
The full file from steipete/CodexBar at commit dd7dca4, republished under its MIT licence (© steipete). 387 words, ~1,231 tokens.
.claude/skills/qa-test/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.Use for live provider testing, release smoke tests, menu verification, or debugging “provider works/fails” reports.
CodexBar.app/Contents/Helpers/CodexBarCLI.CodexBar.app/Contents/MacOS/codexbar; that is the app binary and may appear to hang as a CLI.env, set, or secret regex dumps.$one-password for secrets: all op commands inside one persistent tmux session, service account first, no raw secret output.KeychainNoUIQuery-safe tests unless the user explicitly requested live UI.Run the bundled script:
.agents/skills/qa-test/scripts/live_provider_matrix.sh --enabledUseful modes:
.agents/skills/qa-test/scripts/live_provider_matrix.sh --provider all
.agents/skills/qa-test/scripts/live_provider_matrix.sh --providers openai,zai,deepseek
.agents/skills/qa-test/scripts/live_provider_matrix.sh --defaultInterpretation:
--enabled asks CodexBarCLI config providers for enabled providers, honoring CODEXBAR_CONFIG and default toggles.--default runs the app-facing default command with no provider override.--provider all forces every registered provider and is expected to fail for providers without sessions/keys.--enabled and --default clean; --provider all is a discovery/triage tool.Validate config:
CodexBar.app/Contents/Helpers/CodexBarCLI config validate
stat -f '%Lp %N' "$HOME/.codexbar/config.json"Redact config shape:
jq '(.providers // []) |= map(.apiKey = (if .apiKey then "<redacted>" else .apiKey end) |
.secretKey = (if .secretKey then "<redacted>" else .secretKey end) |
.cookieHeader = (if .cookieHeader then "<redacted>" else .cookieHeader end) |
(if .id == "stepfun" and has("region") then .region = "<redacted>" else . end) |
.tokenAccounts = (if .tokenAccounts then (.tokenAccounts | .accounts = (.accounts | map(.token = "<redacted>"))) else .tokenAccounts end))' \
"$HOME/.codexbar/config.json"Before editing config, make a backup:
cp "$HOME/.codexbar/config.json" "$HOME/.codexbar/config.pre-qa-$(date +%Y%m%d%H%M%S).json"
chmod 600 "$HOME/.codexbar"/config.pre-qa-*.jsonUse Peekaboo after CLI checks:
pkill -x CodexBar || pkill -f 'CodexBar.app/Contents/MacOS/CodexBar' || true
open -n "$PWD/CodexBar.app"
peekaboo menu list-all --json | rg -i 'codexbar'
peekaboo menu click-extra --title codexbar-merged --json
screencapture -x /tmp/codexbar-live-menu.pngCrop top-right menu if needed:
sips --cropToHeightWidth 900 340 --cropOffset 20 2650 /tmp/codexbar-live-menu.png \
--out /tmp/codexbar-live-menu-crop.png >/dev/nullVerify visually with view_image. Confirm provider tabs/rows match enabled config and no failing provider dominates the first screen.
Use $browser-use only when a logged-in dashboard, API key page, or provider docs need browser/profile state.
Existing Chrome path:
mcporter call chrome-devtools.list_pages --args '{}' --output text
mcporter call chrome-devtools.navigate_page --args '{"url":"https://provider.example"}' --output text
mcporter call chrome-devtools.take_snapshot --args '{}' --output textIf Browser Use is unavailable, say so and use web search for public official docs; do not substitute isolated Playwright for login/profile-dependent pages.
CHANGELOG.md.make check, $autoreview, and live CLI proof before landing.OPENAI_API_KEY values can fail legacy credit-balance fallback with 403.© steipete, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 3 other files (scripts, references) in .agents/skills/qa-test of steipete/CodexBar.
Open the folder on GitHubat commit dd7dca4
CodexBar Live QA 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 |
|---|---|---|---|---|---|---|
| CodexBar Live QA this skillsteipete/CodexBar | 22k | — | ~1.2k | Automated safety check: Pass | MIT | |
| Senpi Agent QA Harnesscode-yeongyu/senpi | 470 | — | ~2.7k | Automated safety check: Notes | MIT | |
| tmux Real User TestingQwenLM/qwen-code | 28k | — | ~2.3k | Automated safety check: Pass | Apache-2.0 | |
| Verify Omnigent End-to-Endomnigent-ai/omnigent | 11k | — | ~1.6k | Automated safety check: Pass | Apache-2.0 | |
| Tmux Manual QA Workercode-yeongyu/senpi | 470 | — | ~1.6k | Automated safety check: Pass | MIT | |
| Acceptance Evidence for Deliverieslobehub/lobehub | 83k | — | ~9.7k | Automated safety check: Pass | Apache-2.0 |
code-yeongyu/senpi
Checks changes to the senpi coding agent by driving the real CLI from source in an isolated sandbox, over RPC, terminal UI, mock model and CLI smoke channels.
QwenLM/qwen-code
Drives Qwen Code in a real tmux session the way a user would and saves a readable step-by-step transcript of each screen for maintainers to review.
omnigent-ai/omnigent
Spins up an isolated Omnigent server, runner and mock model to prove a user-facing behavior or bug fix with recorded evidence instead of reasoning from code.
code-yeongyu/senpi
Runs one manual QA scenario for the todo continuation feature in the real ./pi-test.sh CLI inside tmux, captures scrollback and checks a deterministic count marker.
lobehub/lobehub
Verifies a delivery end to end by driving the real product on a CLI, web, desktop or iOS Simulator surface, capturing evidence and publishing a round with the lh CLI.
code-yeongyu/oh-my-openagent
Tests the omo Codex plugin in an isolated CODEX_HOME with a local mock model, proving hooks fired through app-server notifications without touching ~/.codex.
steipete/CodexBar
CodexBar read. Provider usage, limits, credits, config health. JSON. No writes.
steipete/CodexBar
Releases a signed, notarized CodexBar build: confirms the changelog, resolves signing credentials from 1Password, runs the release script in tmux, and updates the Sparkle appcast and Homebrew tap.
Works with
Categories
Runs live QA for the CodexBar app: provider usage matrix checks through its packaged CLI, config validation and menu checks, with 1Password-backed credentials handled safely. The skill is for live provider testing, release smoke tests, menu verification and debugging reports that a provider works or fails. The agent works from the CodexBar checkout and uses the packaged `CodexBarCLI` helper rather than the app binary, which can look like it hangs as a CLI.
CodexBar Live QA fits situations like: running a release smoke test across CodexBar's providers; debugging a report that one provider stopped working; validating CodexBar's config file before and after a change; verifying the menu bar output of the app.
Run `npx skills add steipete/CodexBar --skill qa-test -a claude-code`. Or copy the skill folder (.agents/skills/qa-test in steipete/CodexBar) into .claude/skills/qa-test in your project. Claude Code loads it when a task matches its description.
Run `npx skills add steipete/CodexBar --skill qa-test -a codex`. Or copy the skill folder (.agents/skills/qa-test in steipete/CodexBar) into .agents/skills/qa-test in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add steipete/CodexBar --skill qa-test -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/qa-test, .gemini/skills/qa-test, .github/skills/qa-test and .opencode/skills/qa-test in your project.
Going by SKILL.md and its folder, CodexBar Live QA needs a shell for the scripts in its folder, the command-line tools its instructions call (jq, rg and make) and credentials named OPENAI_API_KEY. Our summary lists: The CodexBar repository checkout with `CodexBar.app` and its packaged CLI; Peekaboo, for menu checks; 1Password and tmux, when credentials are needed.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
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. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
CodexBar Live QA is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 1.2k tokens (SKILL.md is roughly 4.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 189 tokens, read only when the agent opens those files.
Skills that share tags, products or a category with CodexBar Live QA: Senpi Agent QA Harness (code-yeongyu/senpi, 470 stars), tmux Real User Testing (QwenLM/qwen-code, 28k stars), Verify Omnigent End-to-End (omnigent-ai/omnigent, 11k stars) and Tmux Manual QA Worker (code-yeongyu/senpi, 470 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
steipete (a GitHub user) maintains it in steipete/CodexBar, which has 22,250 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on October 7, 2026.
Source: steipete/CodexBar on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.