Back to skill
v1.0.0

Add Music To Video Online

BenignClawScan verdict for this skill. Analyzed Apr 30, 2026, 10:27 PM.

Analysis

This looks like a purpose-aligned online video editing skill, but it sends your media and editing instructions to NemoVideo's cloud API using a service token and remote session.

GuidanceUse this skill only with videos, audio, and prompts you are comfortable sending to NemoVideo's cloud service. Keep NEMO_TOKEN private, be aware that exports may use credits or require plan access, and consider verifying the provider separately because the registry lists no source or homepage.

Findings (8)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Agent Goal Hijack
SeverityLowConfidenceMediumStatusNote
SKILL.md
The backend responds as if there's a visual interface. Map its instructions to API calls: - "click" or "点击" → execute

Backend responses are used to drive subsequent API actions. This is purpose-aligned for a remote editing workflow, but it means external service instructions can influence the agent's action sequence.

User impactThe external editing service may guide parts of the workflow after you ask for edits.
RecommendationUse the skill for its intended video-editing tasks and review final actions such as export/download, especially if credits or account limits matter.
Tool Misuse and Exploitation
SeverityLowConfidenceHighStatusNote
SKILL.md
| `/api/render/proxy/lambda` | POST | Start export. Body: `{"id":"render_<ts>","sessionId":"<sid>","draft":<json>,"output":{"format":"mp4","quality":"high"}}`. Poll status every 30s. |

The skill can start remote render/export jobs and poll status. These API operations are central to the stated purpose, but they may consume service credits or create remote jobs.

User impactYour requested edits and exports are performed through cloud API operations rather than locally.
RecommendationConfirm that you want to upload and export before using it with important or private media.
Agentic Supply Chain Vulnerabilities
SeverityLowConfidenceHighStatusNote
metadata
Source: unknown
Homepage: none

The registry does not provide a source repository or homepage to verify the publisher or service provenance. There is no local code or install step, so this is a provenance note rather than a concrete unsafe install issue.

User impactYou have limited registry-provided information for independently verifying who maintains the skill.
RecommendationUse extra caution with private media and verify the NemoVideo service separately if provenance matters to you.
Cascading Failures
SeverityInfoConfidenceHighStatusNote
SKILL.md
The session token carries render job IDs, so closing the tab before completion orphans the job.

Remote render jobs can continue or become orphaned if the session is interrupted. This is a bounded cloud-job lifecycle issue rather than evidence of broader propagation.

User impactAn interrupted export may leave a remote render job unfinished or orphaned.
RecommendationWait for exports to complete and check status before closing the session when possible.
Human-Agent Trust Exploitation
SeverityInfoConfidenceMediumStatusNote
SKILL.md
Tell the user you're ready. Keep the technical details out of the chat.

The instruction favors a simplified user experience and could reduce visibility into backend token/session setup, although the same artifact does disclose cloud rendering and API use.

User impactYou may not see every backend connection or session detail during normal use.
RecommendationUnderstand that the skill connects to NemoVideo's cloud service even if the chat only shows brief status messages.
Permission boundary

Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.

Identity and Privilege Abuse
SeverityLowConfidenceHighStatusNote
SKILL.md
If `NEMO_TOKEN` is in the environment, use it directly and create a session. Otherwise, acquire a free starter token

The skill uses a provider token or obtains an anonymous token for the NemoVideo service. This is expected for the integration and no unrelated credential use is shown.

User impactThe skill will authenticate to NemoVideo using your token or an anonymous starter token.
RecommendationKeep NEMO_TOKEN private and use a token intended only for this service.
Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Memory and Context Poisoning
SeverityLowConfidenceHighStatusNote
SKILL.md
| `/api/state/nemo_agent/me/<sid>/latest` | GET | Fetch current timeline state (`draft`, `video_infos`, `generated_media`). |

The service keeps session timeline state and generated media metadata. This is expected for cloud editing, but it is persistent remote context that may contain information about the user's project.

User impactYour editing session and media-related state may exist on the remote service while the project is active.
RecommendationAvoid uploading sensitive videos or audio unless you are comfortable with the provider handling the project state.
Insecure Inter-Agent Communication
SeverityLowConfidenceHighStatusNote
SKILL.md
| `/run_sse` | POST | Send a user message. Body includes `app_name`, `session_id`, `new_message`. Stream response with `Accept: text/event-stream`. Timeout: 15 min. |

User editing instructions are sent to a remote backend session over an SSE workflow. This is disclosed and purpose-aligned, but it means prompts and media workflow data leave the local chat.

User impactYour editing requests are processed by NemoVideo's backend, not only by the local agent.
RecommendationDo not include unrelated secrets or sensitive personal details in editing prompts.