Editor Anup Sagar
Analysis
This instruction-only cloud video editor is coherent with its purpose, but it uploads user media to an external Nemo backend and uses a bearer token/session state.
Findings (6)
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.
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.
When a user first opens this skill, connect to the processing backend automatically... POST to `https://mega-api-prod.nemovideo.ai/api/auth/anonymous-token`... Create a session
The skill directs the agent to make external API calls automatically and create a backend session; this is purpose-aligned for cloud editing, but users should know it happens.
Source: unknown; Homepage: none
The skill has limited provenance information while relying on an external cloud backend; this is not malicious by itself, but it reduces independent verification.
closing the tab before completion orphans the job
The skill states that cloud render jobs may continue without the active user session; this is normal for cloud rendering but is persistent background behavior.
Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.
Include `Authorization: Bearer <NEMO_TOKEN>` and all attribution headers on every request
The skill uses a bearer token for all backend requests, including uploads, state, credits, and export operations; this credential is declared and aligned with the service purpose.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
Store the returned `session_id` for all subsequent requests
The skill maintains session context across requests; this is necessary for editing workflows, but it means project state persists beyond a single message.
`/run_sse` | POST | Send a user message... Stream response with `Accept: text/event-stream`
User messages and editing instructions are sent to an external streaming backend/agent; this is expected for the cloud editor but is a data-boundary users should notice.
