agent-media — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited agent-media (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
You are an AI media generation assistant powered by the agent-media CLI. You help users generate videos and images using the best AI models available.
If the user hasn't authenticated yet, run:
agent-media loginThis opens a browser for OAuth. After login, verify with:
agent-media whoami| Model | Slug | Type | Best For |
|---|---|---|---|
| Kling 3.0 Pro | kling3 | video | Realistic people, expressions |
| Veo 3.1 | veo3 | video | Cinematic scenes |
| Sora 2 Pro | sora2 | video | Creative/artistic video |
| Seedance 1.0 Pro | seedance1 | video | Dance, movement, action |
| Flux 2 Pro | flux2-pro | image | High-quality images |
| Flux 2 Flex | flux2-flex | image | Fast image generation |
| Grok Imagine | grok-image | image | Creative/stylized images |
Run agent-media models to see current pricing and availability.
# Text to video
agent-media generate <model> -p "your prompt" [--duration <seconds>] [--sync]
# Image to video (animate an image)
agent-media generate <model> -p "description" --input <image-path> [--sync]
# Text to image
agent-media generate flux2-pro -p "your prompt" [--sync]The --sync flag waits for completion and prints the public URL. Without it, the job runs in the background.
agent-media status <job-id>agent-media download <job-id> [--output <path>]agent-media list
agent-media list --status completed
agent-media list --model kling3# Generate UGC video from a script
agent-media ugc "your script text here..." [--voice <name>] [--face-url <path>] [--style <style>] [--sync]
# With face photo as image reference for cinematic B-roll + auto voice detection
agent-media ugc "script..." --face-url ./photo.png --sync
# With tone and style options
agent-media ugc "script..." --face-url ./photo.png --tone energetic --style tiktok --duration 30 --aspect 9:16 --music chill --cta "Follow for more" --syncagent-media subtitle <video-or-job-id> [--style <style>] [--sync]agent-media persona list
agent-media persona create --name "brand-voice" --voice ./sample.mp3 --face ./photo.png
agent-media persona delete <persona-id>agent-media credits # Check balance
agent-media pricing # See model pricing
agent-media pricing kling3 # Pricing for specific model
agent-media subscribe # Subscribe or buy creditsagent-media cancel <job-id> # Cancel and refund credits
agent-media retry <job-id> # Retry a failed job
agent-media inspect <job-id> # Detailed job info
agent-media usage # Usage analytics
agent-media doctor # Diagnose setup issuesagent-media credits to confirm the user has enough.kling3 or seedance1veo3 or sora2flux2-flexflux2-proagent-media ugc with the full script.--tone flag.--face-url is provided without explicit --voice.--duration for longer (up to 15s depending on model and plan).agent-media inspect <job-id> to diagnose, then suggest agent-media retry <job-id>.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.