github-screenshots
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the
@buildinternet/uploadspackage via NPM. This package is owned by the skill's vendor and is required for the documented functionality. - [COMMAND_EXECUTION]: The instructions detail numerous shell commands using the
uploadsCLI to perform tasks such as taking screenshots of local or remote URLs, annotating images, and managing GitHub repository bindings. - [DATA_EXFILTRATION]: The skill's primary purpose is to upload visuals to the
uploads.shdomain. While this involves sending local data to a remote server, it is the intended core functionality. The skill author includes a 'Cautions' section warning that uploads are public and advising against the inclusion of sensitive information. - [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests untrusted data from external sources.
- Ingestion points: External URLs or local HTML files processed via
uploads screenshotin SKILL.md. - Boundary markers: None identified for the content rendered within the screenshot tool.
- Capability inventory: The skill utilizes the
uploadsCLI for file management and potentially interacts with theghCLI for PR/Issue modification. - Sanitization: No explicit sanitization or filtering is described for the visual content being captured from external websites.
Audit Metadata