appshots-cli

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill's primary entry point (SKILL.md) functions as a redirect, instructing the AI agent to tell the user to install a more comprehensive collection of automation skills from the author's public repository using npx skills add truongduy2611/app-screenshots-cli-skills.
  • [PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface through its AI translation features (translate all and translate set-prompt). These tools allow external data (screenshot text) and user-defined context prompts to be sent to an AI translation engine.
  • Ingestion points: Processes text overlays from screenshot designs and custom context strings for translation.
  • Boundary markers: Not explicitly defined in the CLI code; boundaries are expected to be managed by the host application's server-side logic.
  • Capability inventory: The tool has the ability to read local image files, write exported screenshots to the disk, and communicate with the local host via HTTP.
  • Sanitization: The CLI client does not perform local sanitization or validation of the text content sent to the translation interface.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 02:49 AM
Security Audit — agent-trust-hub — appshots-cli