wordpress-ai-image-optimizer
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted media and metadata from an external WordPress site. This creates an attack surface where malformed files or malicious metadata could attempt to influence the AI agent's behavior or exploit local processing tools.
- Evidence: Found in SKILL.md: 'Downloads images from WordPress to local workspace... processes images locally.'
- Capability Tier: High (The skill can write to the site using 'respira_update_page', 'respira_update_post', and 'respira_sideload_image').
- [COMMAND_EXECUTION]: The skill relies on executing local command-line utilities (for compression and conversion) on files sourced from a remote network location.
- Evidence: Described in SKILL.md under 'What This Skill Does' and 'Requirements' (mentions AI code editor with image processing tools).
Audit Metadata