sfd-video-captions
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No malicious instruction patterns or attempts to override agent behavior were detected in the skill instructions or scripts.
- [DATA_EXFILTRATION]: The skill performs network requests to ZapCap (api.zapcap.ai) and VidJutsu (api.vidjutsu.ai) solely for video processing and CDN uploads. No unauthorized transmission of sensitive system data or credentials was found.
- [CREDENTIALS_UNSAFE]: The skill adheres to security best practices by using environment variables (ZAPCAP_API_KEY and VIDJUTSU_API_KEY) for authentication. No hardcoded secrets or sensitive files are accessed.
- [COMMAND_EXECUTION]: The script utilizes the Bun runtime for standard file I/O and networking. It does not invoke system shells or execute arbitrary external commands.
- [REMOTE_CODE_EXECUTION]: No patterns indicating the download and execution of remote scripts or unverified dependencies were identified.
- [EXTERNAL_DOWNLOADS]: The skill downloads processed video content from the ZapCap service to a user-specified output path as part of its core functionality.
Audit Metadata