video-summarization
Enables efficient video summarization by leveraging VLM NIM and LVS microservices for both short and long videos.
Install this skill
Security score
The video-summarization skill was audited on May 16, 2026 and we found 28 security issues across 4 threat categories, including 9 high-severity. Review the findings below before installing.
Categories Tested
Security Issues
Template literal with variable interpolation in command context
| 60 | | `< 60s` (short) | **VLM NIM** (OpenAI-compatible) | `POST ${VLM_BASE_URL}/v1/chat/completions` | |
Template literal with variable interpolation in command context
| 61 | | `>= 60s` (long), LVS available | **LVS microservice** | `POST ${LVS_BACKEND_URL}/summarize` | |
Template literal with variable interpolation in command context
| 62 | | `>= 60s`, LVS **not** reachable | **VLM NIM** + tell the user | `POST ${VLM_BASE_URL}/v1/chat/completions` | |
Template literal with variable interpolation in command context
| 78 | ```bash |
Template literal with variable interpolation in command context
| 108 | - VLM NIM: `${VLM_BASE_URL}` — default `http://localhost:30082` |
Template literal with variable interpolation in command context
| 109 | - LVS MS: `${LVS_BACKEND_URL}` — default `http://localhost:38111` |
Template literal with variable interpolation in command context
| 121 | **Model name:** read `${VLM_NAME}` (default `nvidia/cosmos-reason2-8b`). |
Template literal with variable interpolation in command context
| 134 | ```bash |
Template literal with variable interpolation in command context
| 202 | ```bash |
Template literal with variable interpolation in command context
| 245 | Full scenario/events collection walk-through lives in [`references/hitl-prompts.md`](references/hitl-prompts.md). Always run this step before calling LVS. |
Template literal with variable interpolation in command context
| 280 | ```bash |
Template literal with variable interpolation in command context
| 311 | ```bash |
Curl to non-GitHub URL
| 79 | curl -sf --max-time 5 "http://${HOST_IP}:8000/docs" >/dev/null \ |
Curl to non-GitHub URL
| 80 | && curl -sf --max-time 5 "http://${HOST_IP}:38111/v1/ready" >/dev/null |
Curl to non-GitHub URL
| 205 | curl -s -X POST "${VLM_BASE_URL:-http://localhost:30082}/v1/chat/completions" \ |
Curl to non-GitHub URL
| 247 | curl -s -X POST "${LVS_BACKEND_URL:-http://localhost:38111}/summarize" \ |
Curl to non-GitHub URL
| 290 | curl -s -X POST "${VLM_BASE_URL:-http://localhost:30082}/v1/chat/completions" \ |
Access to .env file
| 33 | [`references/lvs.env.example`](references/lvs.env.example). |
External URL reference
| 79 | curl -sf --max-time 5 "http://${HOST_IP}:8000/docs" >/dev/null \ |
External URL reference
| 80 | && curl -sf --max-time 5 "http://${HOST_IP}:38111/v1/ready" >/dev/null |
External URL reference
| 108 | - VLM NIM: `${VLM_BASE_URL}` — default `http://localhost:30082` |
External URL reference
| 109 | - LVS MS: `${LVS_BACKEND_URL}` — default `http://localhost:38111` |
External URL reference
| 137 | "${VLM_BASE_URL:-http://localhost:30082}/v1/models") |
External URL reference
| 141 | LVS=${LVS_BACKEND_URL:-http://localhost:38111} |
External URL reference
| 205 | curl -s -X POST "${VLM_BASE_URL:-http://localhost:30082}/v1/chat/completions" \ |
External URL reference
| 247 | curl -s -X POST "${LVS_BACKEND_URL:-http://localhost:38111}/summarize" \ |
External URL reference
| 290 | curl -s -X POST "${VLM_BASE_URL:-http://localhost:30082}/v1/chat/completions" \ |
External URL reference
| 312 | LVS=${LVS_BACKEND_URL:-http://localhost:38111} |