本地 Ollama 文生图工具。使用 x/z-image-turbo 模型生成图片，分辨率 1024x1024，每张图约耗时 90 秒。当用户说"生成图片"、"文生图"、"画一张图"、或提交中文提示词生成图像时使用。

v1.0.0

⭐ 0· 180·0 current·0 all-time

byThomas@jiafeimao-gjf

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for jiafeimao-gjf/ollama-t2i.

Previewing Install & Setup.

Prompt PreviewInstall & Setup

Install the skill "本地 Ollama 文生图工具。使用 x/z-image-turbo 模型生成图片，分辨率 1024x1024，每张图约耗时 90 秒。当用户说"生成图片"、"文生图"、"画一张图"、或提交中文提示词生成图像时使用。" (jiafeimao-gjf/ollama-t2i) from ClawHub.
Skill page: https://clawhub.ai/jiafeimao-gjf/ollama-t2i
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install ollama-t2i

ClawHub CLI

Package manager switcher

npx clawhub@latest install ollama-t2i

Security Scan

VirusTotal

Pending

View report →

OpenClaw

Benign

high confidence

✓

Purpose & Capability

Name/description (local Ollama text→image using x/z-image-turbo) matches the code and instructions. No unrelated binaries, env vars, or external services are requested.

ℹ

Instruction Scope

SKILL.md and the script limit network activity to http://localhost:11434 (local Ollama). The script writes PNG files to an images/ directory. Note: SKILL.md does not declare the Python requests dependency required by scripts/t2i.py, and the usage example references an absolute path under ~/.openclaw which may not match every install location.

✓

Install Mechanism

No install spec — instruction-only plus a small helper script. Nothing is downloaded or installed by the skill itself.

✓

Credentials

The skill requires no environment variables, credentials, or config paths. Prompts are sent only to the local Ollama endpoint; no secrets are requested or stored by the skill.

✓

Persistence & Privilege

always:false and no code that modifies other skills or system-wide agent settings. The script only creates/uses its own output directory.

Assessment

What to check before installing/use: 1) Ensure you run a trusted Ollama instance at http://localhost:11434 and that you have pulled the x/z-image-turbo model as instructed; if your Ollama is configured to forward requests externally, prompts could leave your machine — verify Ollama's network/telemetry settings. 2) The script uses the Python requests package; install it if missing (pip install requests). 3) The example uses an absolute skill path (~/.openclaw/...), adjust if your skill is installed elsewhere. 4) The script saves base64-decoded images into an images/ directory (filenames randomized) — verify disk location and free space. 5) Minor bug: the printed 'file size' is the length of the base64 string, not exact byte size of the PNG. Overall the skill appears to do only local image generation and file writes and does not request credentials or contact remote services, so it is coherent with its description.

Like a lobster shell, security has layers — review code before you run it.

latestvk978vptgwyfabv08rxzrkwyt2983dwpv

180downloads

0stars

1versions

Updated 1mo ago

v1.0.0

MIT-0

Ollama 文生图 (Text-to-Image)

环境要求

Ollama 服务运行于 http://localhost:11434
模型：x/z-image-turbo（需提前拉取 ollama pull x/z-image-turbo）
输出目录：images/（自动创建）

使用方法

Python 脚本

运行以下命令生成图片：

python3 ~/.openclaw/openclaw/skills/ollama-t2i/scripts/t2i.py "你的中文提示词"

单次生成

python3 ~/.openclaw/openclaw/skills/ollama-t2i/scripts/t2i.py "五名青年男子并排站立，着装休闲，多背双肩包..."

批量生成

编辑脚本内的 prompts 列表，或传多个参数：

python3 ~/.openclaw/openclaw/skills/ollama-t2i/scripts/t2i.py "提示词1" "提示词2"

输出

文件名格式：YYYY-MM-DD_HH-MM_image_<随机数>.png
保存路径：images/
包含请求耗时统计

代码位置

scripts/t2i.py

Comments

Loading comments...