v1.0.0

Professional Ai Image

BenignClawScan verdict for this skill. Analyzed Apr 30, 2026, 10:23 PM.

Analysis

This skill appears purpose-aligned for cloud image-to-video enhancement, but it uses a NemoVideo token and sends user media and session data to an external cloud API.

GuidanceInstall this only if you are comfortable sending your images, prompts, and generated media workflow to the NemoVideo cloud API. Protect NEMO_TOKEN like a credential, avoid uploading sensitive personal media unless you trust the provider, and ask for confirmation before uploads or exports if you want stricter control.

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

SeverityLowConfidenceHighStatusNote

SKILL.md

Backend Response Translation: Backend says 'click [button]' / 'Export button' -> You do 'Execute via API' / 'Execute export workflow'

The skill makes remote backend responses drive follow-up API actions. This is limited to the intended video workflow, but users should know the backend can guide the agent's next steps.

User impactA response from the cloud service may cause the agent to continue the workflow, such as querying state or exporting, without restating every technical step to the user.

RecommendationUse the skill for media tasks you intend to process, and ask the agent to confirm before export or other account-affecting actions if you want tighter control.

Tool Misuse and Exploitation

SeverityLowConfidenceHighStatusNote

SKILL.md

Upload: POST /api/upload-video/nemo_agent/me/<sid> — file: multipart -F 'files=@/path', or URL: {'urls':['<url>'],'source_type':'url'}

The skill can upload user-provided files or URLs to the backend and then start render/export workflows. This is central to the stated purpose, but it is a meaningful external action.

User impactFiles or URLs you provide may be sent to the cloud rendering service for processing.

RecommendationOnly provide files and URLs you intend to upload to the NemoVideo backend, especially if they contain private images, audio, or video.

Agentic Supply Chain Vulnerabilities

SeverityInfoConfidenceHighStatusNote

metadata

Source: unknown; Homepage: none

The skill has no installable code or dependencies, but the registry metadata does not provide a source or homepage for verifying the external service integration.

User impactIt may be harder to independently verify who maintains the skill or where to review service documentation.

RecommendationConfirm you trust the skill publisher and the NemoVideo API endpoint before sending sensitive media.

Human-Agent Trust Exploitation

SeverityLowConfidenceHighStatusNote

SKILL.md

Tell the user you're ready. Keep the technical details out of the chat.

The skill prefers a simplified user experience and does not require the agent to disclose backend token/session details in chat. Cloud processing is otherwise disclosed, so this is a transparency note rather than deceptive behavior.

User impactYou may not see the exact backend endpoint or token/session steps during normal use.

RecommendationIf the media is sensitive, ask the agent to explain where files are sent and what account or token is being used before uploading.

Rogue Agents

SeverityLowConfidenceHighStatusNote

SKILL.md

The session token carries render job IDs, so closing the tab before completion orphans the job.

Cloud render jobs may continue independently once started. This is expected for rendering and is not evidence of self-propagation or hidden persistence.

User impactA render job you start may continue on the cloud service even if you leave before it finishes.

RecommendationStart exports only when you intend to complete them, and check provider controls if you need to cancel or delete jobs.

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

metadata

Required env vars: NEMO_TOKEN; Primary credential: NEMO_TOKEN

The skill requires a service token and uses it for Bearer authorization. This is expected for the external API integration and no hardcoded credential or unrelated account access is shown.

User impactAnyone with the token may be able to use the associated NemoVideo service access or credits.

RecommendationStore NEMO_TOKEN securely, avoid sharing it in chat, and rotate it if you suspect exposure.

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

Session state: GET /api/state/nemo_agent/me/<sid>/latest — key fields: data.state.draft, data.state.video_infos, data.state.generated_media

The backend keeps session state for drafts, video information, and generated media. This is expected for rendering, but it means task context and media metadata may persist remotely.

User impactYour prompts, draft state, and generated media references may remain in the cloud service session.

RecommendationAvoid uploading highly sensitive media unless you accept the provider's storage and retention practices.

Insecure Inter-Agent Communication

SeverityLowConfidenceHighStatusNote

SKILL.md

Send message (SSE): POST /run_sse — body {'app_name':'nemo_agent','user_id':'me','session_id':'<sid>'...} with Accept: text/event-stream

The skill communicates with a remote agent-like backend over SSE and uses returned messages to continue the workflow. The host and authorization are specified, so this is purpose-aligned rather than hidden.

User impactYour media instructions and workflow state are exchanged with the external NemoVideo backend.

RecommendationUse this only for content you are comfortable sending to that provider, and keep the bearer token private.