skills/gco-translate/SKILL.md
Translation using GitHub Copilot CLI (free GPT-4.1 tier). PREFERRED when Codex quota is low. Use when: (1) User says 'gco translate', 'copilot translate', or '/gco-translate', (2) Translating between languages (Japanese, English, etc.), (3) Translating documentation, comments, or UI strings with zero token cost. Copilot drafts, Claude Code reviews and writes. Falls back to Claude Code if Copilot rate-limited.
npx skillsauth add takazudo/claude-resources gco-translateInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Translation via GitHub Copilot CLI (gco-pure.sh), reviewed and finalized by Claude Code. Uses GPT-4.1 on the free Copilot tier.
GCO_PURE="$HOME/.claude/skills/gco/scripts/gco-pure.sh"
If the file does not exist or is not executable, report "gco-pure.sh not found. Ensure base/copilot-text-tools branch is merged." and fall back to Claude Code translation.
gco-pure.sh passes zero tool flags to Copilot — the model cannot read files or execute code. All file I/O is handled by Claude Code.
Check Copilot's current rate-limit state (informational only):
RATE_CHECK=$(node "$HOME/.claude/scripts/gco-rate-limit.js" check 2>&1)
gco-rate-limit.js check always exits 0. Output is either ok or degraded: … (Copilot in low-cost mode, still usable). Proceed regardless — this is informational context only.
Inline syntax: /gco-translate translate "hello" from en to ja
File syntax: /gco-translate translate path/to/source.md from en to ja
Translate the following text from <source-lang> to <target-lang>.
Context: <what the text is for — technical documentation, UI, blog, etc.>
Tone: <formal/casual/technical>
Rules:
- Preserve technical terms and code references as-is
- Maintain formatting (markdown, HTML tags, etc.)
- Preserve line breaks and paragraph structure
- Use natural phrasing in the target language
Text to translate:
---
<source text>
---
For file translation: Claude reads the source file content first. For content over ~100KB, write it to a temp file and pass as the second positional arg to gco-pure.sh.
GCO_PURE="$HOME/.claude/skills/gco/scripts/gco-pure.sh"
LOGDIR=$(node "$HOME/.claude/scripts/get-logdir.js")
mkdir -p "$LOGDIR"
DATETIME=$(date +%Y%m%d_%H%M%S)
DRAFT_LOG="$LOGDIR/${DATETIME}-gco-translate-draft.md"
STDERR_LOG="$LOGDIR/${DATETIME}-gco-translate-draft-stderr.log"
# Guard: ensure gco-pure.sh exists — if missing, skip to Step 5 (fallback)
if [ ! -x "$GCO_PURE" ]; then
echo "gco-pure.sh not found or not executable: $GCO_PURE" >&2
exit 1 # caller (Claude) catches this and proceeds to Step 5
fi
# Inline translation: pipe source text via stdin
printf '%s' "<source text>" \
| bash "$GCO_PURE" "<translation prompt>" \
> "$DRAFT_LOG" \
2>"$STDERR_LOG"
# File translation (preferred for files, avoids shell-escaping issues):
# bash "$GCO_PURE" "<translation prompt>" "<source-file-path>" \
# > "$DRAFT_LOG" \
# 2>"$STDERR_LOG"
gco-pure.sh handles its own 15-minute internal timeout. No outer timeout needed.
After Copilot completes, check output files:
node "$HOME/.claude/scripts/gco-rate-limit.js" check-output \
"$DRAFT_LOG" \
"$STDERR_LOG"
RATE_OUT_EXIT=$?
If RATE_OUT_EXIT is non-zero (rate limit pattern detected), proceed to Step 5: Fallback.
Then check if draft output is empty:
DRAFT_CONTENT=$(cat "$DRAFT_LOG" 2>/dev/null)
If DRAFT_CONTENT is empty or blank, proceed to Step 5: Fallback. Do not continue to Step 6 with an empty draft.
If gco-pure.sh is not found, Copilot times out, produces no output, or rate limit is detected:
For inline translation: output the final translation to stdout.
For file translation: Claude writes the output file directly. Copilot never touches the filesystem.
Present the translation to the user. For file translation, show the output file path. Include the draft log path for debugging.
gco-pure.sh has an internal 15-minute (900s) timeout — no outer wrapper neededgco-pure.sh runs with --available-tools (empty list) — zero tool access to Copilotgco-pure.sh "<prompt>" "<file-path>" form (second positional arg) instead of stdin to avoid shell-escaping issues$LOGDIR/${DATETIME}-gco-translate-draft.md for debugging even in fallback scenariosdevelopment
Link Claude Code skill names mentioned in a CodeGrid article (data/{series}/{n}.md) to the author's public claude-resources repo, pinned to the latest commit hash so links don't rot. Use when: (1) user says 'linkify cc resources', 'link the skills', 'link skill names', or invokes /dev-linkify-cc-resources; (2) editing a CodeGrid article that mentions `/commits`, `/pr-complete`, `/skill-creator` or other Claude Code skills and they should point to claude-resources. Only links skills that actually exist in the public repo; skips hypothetical examples and code blocks.
development
Second opinion from Claude Opus on a plan or approach. Use when: (1) Planning phase of /big-plan needs a higher-quality review than /codex-2nd / /gco-2nd / /gcoc-2nd, (2) User says 'opus 2nd' or 'opus opinion', (3) Wanting Anthropic's larger model to critique a plan. Spawns a general-purpose Agent with model: opus that reads the plan file and returns structured feedback. Anthropic quota — not free.
tools
AI-based testing via subagent + a per-task test-flow skill. Use when the user wants to verify something that mechanical assertions can't fully capture — image recognition, visual size/position comparison, animation smoothness, multi-step manual flows that need AI judgment. Triggers: 'AI-based test', 'AI test', 'visual verify', 'image recognition test', 'manual operation test', 'human-eye check', 'verify visually', 'compare screenshots', 'looks the same', 'looks correct'. The skill's job is to (1) author a focused test-flow skill that captures the exact procedure + verdict criteria, then (2) dispatch a verification subagent via the Agent tool that loads BOTH the test-flow skill AND a browser-driving skill (/verify-ui primary, /headless-browser fallback) so the subagent has clear context and consistent verdicts. NEVER uses `claude -p` — subagent dispatch goes through the Agent tool exclusively.
development
End-of-workflow audit of touched GitHub issues, PRs, and branches via a Sonnet subagent. Use when: (1) /big-plan, /x-as-pr, or /x-wt-teams finishes its main work and needs to verify every touched resource is in the right state (closed when done, kept when ongoing, deleted when dead), (2) User says 'cleanup resources', 'audit cleanup', or 'check what should be closed', (3) A long workflow ends and the manager wants a structured paper trail of what it closed/kept/deleted. Auto-execute by default — the Sonnet agent proposes, the manager (you) executes safe actions and prints a final report.