Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Poster Forge

v1.0.0

Universal image and poster generator with Chinese/English text support. Use when asked to create posters, social media images, cover images, infographics, co...

0· 90·0 current·0 all-time
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description (poster/image generation, Chinese/English text support) align with the included files and runtime requirements. The script implements three engines (AI, HTML, text) and requires Python + Pillow, curl, and optionally Chromium — all reasonable for the stated purpose.
Instruction Scope
SKILL.md instructs the agent to run scripts/generate.py which performs only poster-related tasks. The script downloads fonts from GitHub raw URLs, fetches images from Pollinations.ai (https://image.pollinations.ai), writes temp HTML files and screenshots, and runs headless Chromium. It does not read unrelated system files or expect other credentials. Note: it does send prompts and image requests over the network to external services.
Install Mechanism
No install spec is provided (instruction-only + one script), so nothing is written to system install locations. The script itself uses subprocess calls to curl and Chromium but does not perform any packaged installs or download arbitrary executables. Font downloads are from GitHub raw URLs (public font files).
Credentials
The skill requests no environment variables, no credentials, and no config paths. All external accesses (font raw URLs and Pollinations.ai) are consistent with generating images; there is no unexplained credential access or secret exfiltration.
Persistence & Privilege
always is false and the skill does not request permanent agent presence or modify other skills/configs. It writes temporary files (fonts in the system temp dir and temporary HTML/screenshot files) within its own runtime scope.
Assessment
This skill appears to do what it says, but be aware it performs network operations: it downloads fonts from GitHub raw URLs and sends prompts to Pollinations.ai to generate base images, and it uses headless Chromium to render HTML templates. If you have privacy concerns about sending prompts/images to an external service, run the script in an isolated environment or use the HTML/text engines only (Chromium/PIL) which can operate offline. Also ensure you have Python 3 and Pillow installed, and that running headless Chromium with --no-sandbox is acceptable in your environment. Finally, review licensing for any fonts or third-party services if you plan commercial use.

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

latestvk972nv0fv6gq76kz46vnm0tt6d84ctv0
90downloads
0stars
1versions
Updated 1w ago
v1.0.0
MIT-0

Poster Forge

Universal poster/image generator with 3 engines and auto-fallback.

Engines

EngineHowBest ForReliability
aiPollinations.ai (flux model)Artistic/photo backgrounds~90% (network dependent)
htmlChromium headless screenshotPrecise layouts, text-heavy~99% (needs Chromium)
textPIL solid color + text overlaySimple text posters100%

Default mode is auto: tries AI → HTML → text fallback.

Quick Start

# Simple poster with AI background
python3 scripts/generate.py --title "你的标题" --prompt "watercolor sunset" --output poster.jpg

# Split comparison (HTML engine, no network needed)
python3 scripts/generate.py --title "A vs B" --mode html --template split \
  --left-label "方案A" --right-label "方案B" \
  --left-items "优点1|优点2" --right-items "优点1|优点2" --output compare.jpg

# Tutorial card
python3 scripts/generate.py --title "教程标题" --mode html --template tutorial \
  --code-content "print('hello')" --tagline "底部提示语" --output tutorial.jpg

Parameters

Core

  • --title (required): Main title
  • --subtitle: Secondary text
  • --tagline: Bottom tagline
  • --mode: ai | html | text | auto (default: auto)
  • --output: Output file path

Templates (HTML engine)

  • --template: split | gradient | card | tutorial

Size presets

  • --preset: xiaohongshu (1080×1440) | wechat (900×500) | instagram (1080×1080) | twitter (1200×675) | a4 (2480×3508)
  • --width / --height: Custom dimensions

Split template

  • --left-label / --right-label: Side labels
  • --left-items / --right-items: Pipe-separated content items
  • --left-color / --right-color: R,G,B colors

Text overlay

  • --text-position: bottom | center | top | none
  • --font-title-size / --font-sub-size: Font sizes
  • --no-overlay: Skip PIL text overlay (when HTML already has text)

AI engine

  • --prompt: English description for Pollinations.ai

Workflow for Agents

  1. Decide content and style
  2. Choose mode: auto for flexibility, html for reliability, ai for aesthetics
  3. Run scripts/generate.py with appropriate args
  4. If using html mode with --template, add --no-overlay (text is already in the HTML)
  5. Send output image to user

Requirements

  • Python 3 + Pillow (PIL)
  • curl (for AI engine font download)
  • Chromium/Chrome (for HTML engine, optional)

Comments

Loading comments...