菜单
Choose CLI tools and safe patterns before repo work. Install checklist, bounded search/read commands, and confirmation gates. Use when navigating the codebase, running shell commands, or setting up agent tooling on Windows, WSL2, or macOS.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
Update CHANGELOG.md and package.json via compiled release_bump harness, then commit/push only when the user asks. Use for /release, changelog updates, or shipping gallery versions.
Works in image-scoring-gallery on electron/db.ts contract, IPC, apiService, React/Vite, and alignment with sibling image-scoring-backend schema. Use for desktop app bugs, database.engine modes, API URL/port from webui.lock, or TypeScript changes under electron/ or src/. Triggers include Electron gallery, db.ts, IPC, apiService, config.api, PostgreSQL vs api engine, and Vite renderer.
React component patterns, CSS Modules styling, design tokens, virtualization, and UX constitution for the Driftara Gallery renderer. Use for src/ UI work, styling, FilterPanel/GalleryGrid/ImageViewer, or design:check—not electron/db unless coordinated with gallery-electron-ts.
Verify implementation against spec AC-n criteria via compiled harness. Use after /implement or /test-and-fix, before /pr-ready. Parses ACs and emits the report; LLM assigns Verified/Failed/Unknown when evidence is not a clean command exit.
Use before claiming work is complete, fixed, passing, ready to commit, or ready for PR. Runs scripts/agent_skills/verification_before_completion.py for the claim→proof catalog; LLM interprets output. Never upgrade incomplete verification.
Driftara Gallery MCP — is-ui-* router-first; backend triage via sibling is-be-mcp search+dispatch.
| name | agent-cli-hub |
| description | Choose CLI tools and safe patterns before repo work. Install checklist, bounded search/read commands, and confirmation gates. Use when navigating the codebase, running shell commands, or setting up agent tooling on Windows, WSL2, or macOS. |
Router for lightweight CLI tooling skills in image-scoring-gallery. Prefer bounded, low-memory commands before broad reads or heavyweight indexers.
Central entry point for agent-safe CLI workflows: which skill to use, how to install tools, how to bound output, and when to ask for confirmation.
| Need | Skill |
|---|---|
| Find files / grep / browse tree | agent-search — rg vs grep vs ast-grep vs fd; optional fff-gallery project MCP |
| git status, diff, gh PR/issue | agent-git-workflows |
| jq/yq on JSON/YAML, API probes | agent-data-config |
| npm lint, tsc, vitest, docker | agent-dev-tooling |
| Windows vs WSL2 choice | agent-platform-tooling |
| MCP vs CLI code intelligence | mcp-code-intelligence |
| Gallery pipeline / DB triage | image-scoring-mcp |
| Fast indexed file search (MCP) | fff — project fff-gallery in .cursor/mcp.json; see AGENTS.md § fff |
| Architecture / cross-module graph | Graphify — skill graphify; soft rule .cursor/rules/graphify.mdc; see AGENTS.md § Graphify |
git, rg, fd, jq, node (+ npm). Install full set via references/install-blocks.md.
Install in order — see references/install-tiers.md:
git, rg, fd, jq, nodeyq, just, mise, …)fzf, semgrep, …)After installing CLI tools, restart Cursor and verify PATH — see references/agent-environment.md.
See references/install-blocks.md — Windows winget, WSL2 apt, macOS Homebrew.
git status --short before editing.npm run …) over ad hoc commands — see .agent/COMMANDS.md.package.json, pyproject.toml, etc.Full list: references/commands-requiring-confirmation.md.
references/windows-wsl-split.md — when Windows native is enough vs WSL2 for sibling backend work.
/mnt/c for heavy repos; clone under ~/src.npm run doctor for config and backend URL probes.git --version && rg --version && node --version && jq --version
npm run doctor # gallery-specific