skills/agentskillexchange/skills/gallery-dl Image Gallery and Collection Downloader/Gen Agent Trust Hub
gallery-dl Image Gallery and Collection Downloader
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation references external content from a well-known public repository.
- Evidence:
https://github.com/mikf/gallery-dlandhttps://raw.githubusercontent.com/mikf/gallery-dl/HEAD/README.rst(SKILL.md). - [INDIRECT_PROMPT_INJECTION]: The skill facilitates downloading content and metadata from various third-party websites, creating a potential surface for indirect prompt injection if the agent interprets unverified external data as instructions.
- Ingestion points: Data extracted from various hosting sites (e.g., Pixiv, Twitter, Reddit) via the tool's core functionality.
- Boundary markers: None provided in the skill instructions to delimit tool output.
- Capability inventory: Command-line execution of the gallery-dl utility and local file system writes.
- Sanitization: No explicit validation or filtering instructions are provided for processed external content.
Audit Metadata