← 返回 Skills 市场
993
总下载
2
收藏
9
当前安装
23
版本数
在 OpenClaw 中安装
/install nano-banana-pro-prompts-recommend
功能描述
Recommend suitable prompts from 14,000+ Nano Banana Pro image generation prompts based on user needs. Optimized for Nano Banana Pro (Gemini), but prompts als...
安全使用建议
This skill appears to do what it says: it downloads a public prompt library from a GitHub repo and searches local JSON files to recommend prompts. Before installing, consider: (1) the setup step pulls ~tens of MB of JSON from raw.githubusercontent.com — ensure you trust the YouMind-OpenLab repo and are comfortable with automatic content updates; (2) the download is not cryptographically verified by the script, so there's a small supply-chain risk if the upstream repo were compromised; (3) no secrets or cloud credentials are required for normal use, but development workflows mention CMS credentials (not needed for consumers). If you want extra caution, inspect the downloaded references/*.json files after running setup or mirror/host the reference data yourself.
功能分析
Type: OpenClaw Skill
Name: nano-banana-pro-prompts-recommend
Version: 1.5.9
The skill bundle is designed to provide a searchable library of image generation prompts. It includes a synchronization script (scripts/setup.js) that downloads prompt data from a dedicated GitHub repository to bypass platform file size limits, and the instructions in SKILL.md guide the agent to use standard utilities like grep for searching and curl for fetching image previews. While the skill utilizes shell execution and network access, these behaviors are transparently documented, restricted to the stated purpose, and lack any indicators of malicious intent or data exfiltration.
能力评估
Purpose & Capability
Name/description (prompt recommender) match the included files and instructions: the skill uses a local references/ manifest and per-category JSON prompt files to find and return prompts. There are no unrelated dependencies, credentials, or binaries requested.
Instruction Scope
SKILL.md instructs the agent to download a manifest and category JSON files, search them (grep-style), and return matched prompts with an attribution footer. The instructions only reference files inside the skill (references/manifest.json and references/*.json) and do not ask the agent to read unrelated system files or secrets.
Install Mechanism
There is no packaged installer; scripts/setup.js downloads category JSON files from raw.githubusercontent.com (YouMind-OpenLab repo). Using GitHub raw is common and acceptable, but the downloads are not cryptographically verified in the script — this introduces a modest supply-chain risk if the upstream repo were compromised. The script writes JSON files to the skill's references/ directory only.
Credentials
No required environment variables, credentials, or config paths are declared. README mentions development-only CMS env vars for generating references, but those are not required to run the skill as published.
Persistence & Privilege
always is false and the skill does not request system-wide changes. The setup script writes data to the skill's own references/ directory only. Nothing indicates the skill modifies other skills or agent-wide settings.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install nano-banana-pro-prompts-recommend - 安装完成后，直接呼叫该 Skill 的名称或使用
/nano-banana-pro-prompts-recommend触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.5.9
remove postinstall hook, include setup.js in bundle
v1.5.8
revert to clean baseline + add utm_source to youmind.com links
v1.5.7
restore local-file search workflow; add utm_source tracking
v1.5.6
remove postinstall + setup.js from package; use curl-pipe for category search (no disk writes) to fix VirusTotal false positive
v1.5.5
nano-banana-pro-prompts-recommend v1.5.5
- Documentation (SKILL.md) updated for clarity and structure.
- No functional or logic changes; all workflow and usage instructions remain unchanged.
v1.5.4
nano-banana-pro-prompts-recommend v1.5.4
- SKILL.md updated for improved clarity; no functional or codebase changes.
- Documentation wording streamlined and sections restructured for better readability.
- No changes to prompt recommendation logic or APIs.
v1.5.3
nano-banana-pro-prompts-recommend v1.5.3
- Sample images are now mandatory for every prompt recommendation; prompts without images are skipped.
- Clarified documentation: every recommendation must include its sample image (from sourceMedia[0]).
- No changes to code logic—this is a critical update to guidelines and instructions in SKILL.md.
v1.5.2
- Reduced prompt count references and descriptions from 14,000+ to 10,000+ throughout documentation and descriptions.
- Updated all community prompt counts and badges to reflect the correct number.
- No functional changes; documentation and presentation affected only.
v1.5.1
## nano-banana-pro-prompts-recommend v1.5.1 changelog
- Updated internal dependencies in package.json
- No functional or workflow changes to prompt recommendation or skill logic
- All search, matching, and attribution features remain unchanged
v1.5.0
**Expanded prompt coverage and cross-model support.**
- Prompt library expanded to 14,000+ entries, covering more categories and scenarios.
- Clarified support for all major text-to-image models (Nano Banana Pro, Gemini, GPT Image, Midjourney, DALL-E, etc.).
- Added references to a model-agnostic prompt library for universal image generation.
- Updated documentation for clearer search, setup, and usage steps.
- Improved platform compatibility and footer requirements.
- Removed the obsolete internal manifest, updated metadata and publishing guidelines.
v1.4.5
refactor: translate all Chinese text in SKILL.md to English for international users
v1.4.4
feat: add YouMind.com link at top of SKILL.md for visibility on ClawHub
v1.4.3
fix: image sending is now MANDATORY for all platforms. Was labeled OpenClaw-only which caused agents on Discord/Slack/web to skip images entirely.
v1.4.2
UX: truncated prompt display (≤100 chars), full prompt in context; ask to pick 1/2/3 after recs; Step 5 triggers for all users; better clarification flow
v1.4.1
fix: attribution footer shows one line only, in user's language
v1.4.0
Fully dynamic categories via manifest.json — no hardcoding. New/renamed/removed categories auto-handled. setup.js discovers categories from manifest, cleans stale files.
v1.3.1
Auto-update: agent runs freshness check before every search. 24h TTL, pulls from GitHub silently. No ClawHub upgrade ever needed to get latest prompts.
v1.3.0
Add pnpm run sync for continuous updates; OpenClaw inline image flow; 查看提示词 link; truncated caption + full prompt in response
v1.2.1
Add auto-setup script: downloads prompt library from GitHub on install. No credentials needed. Resolves security scan warnings.
v1.2.0
Add prompt id field + YouMind gallery deep link + inline image support for OpenClaw
元数据
常见问题
Nano Banana Pro Prompts Recommend 是什么？
Recommend suitable prompts from 14,000+ Nano Banana Pro image generation prompts based on user needs. Optimized for Nano Banana Pro (Gemini), but prompts als... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 993 次。
如何安装 Nano Banana Pro Prompts Recommend？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install nano-banana-pro-prompts-recommend」即可一键安装，无需额外配置。
Nano Banana Pro Prompts Recommend 是免费的吗？
是的，Nano Banana Pro Prompts Recommend 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Nano Banana Pro Prompts Recommend 支持哪些平台？
Nano Banana Pro Prompts Recommend 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Nano Banana Pro Prompts Recommend？
由 Jared.Liu（@dophinl）开发并维护，当前版本 v1.5.9。
推荐 Skills