Exposed secret literal
Critical
- Finding
- File appears to expose a hardcoded API secret or token.
Morfeo UGC Engine
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This skill is coherent for controlling a local Morfeo UGC pipeline, but it embeds an admin API token and can create or publish social-content drafts, so it needs careful review before use.
Before installing, rotate or remove the embedded token and configure it privately with `MORFEO_API_TOKEN`. Treat the skill as able to control your local Morfeo engine, restart its PM2 service, and create social-media drafts; approve those actions manually and verify the external Morfeo Engine code is trusted.
Static analysis
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
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.
What this means
Anyone who sees or reuses the skill may learn the admin token and could control the local Morfeo API if they can reach it.
Why it was flagged
The artifact exposes a reusable bearer token described as the API token, while the registry metadata declares no credential requirement. That token is used for mutation and publish-related endpoints.
Skill content
- **Token**: `morfeo-admin-2026` (o `$MORFEO_API_TOKEN`) - **Auth header**: `Authorization: Bearer morfeo-admin-2026`
Recommendation
Remove the hardcoded token, rotate it, require `MORFEO_API_TOKEN` through a secret store or local configuration, and limit the API to trusted local callers.
What this means
If invoked, the agent can create new content runs, advance generation steps, and create social-media drafts or publication artifacts.
Why it was flagged
The skill documents endpoints that mutate pipeline state and create a publish/draft action. This is aligned with the stated UGC-engine purpose, but it is high-impact account/content control.
Skill content
Crear run nuevo ... POST /api/queue/add ... Avanzar run ... POST /advance ... Publicar en redes (Postiz → borrador) ... POST /publish
Recommendation
Require explicit confirmation before creating runs, advancing irreversible stages, or sending anything to Postiz; show the final video and caption before publish/draft actions.
What this means
The agent could restart the local engine or start a persistent service if the user approves or asks for diagnostics.
Why it was flagged
The diagnostic instructions can restart or start a local Node service. This is expected for managing the engine, but it executes local project code and affects a persistent process.
Skill content
cd ~/clawd/projects/morfeo-engine/api && pm2 restart morfeo-engine-api # Si no existe: pm2 start server.js --name morfeo-engine-api
Recommendation
Run PM2 start/restart commands only with explicit user approval and only after verifying the local repository path is trusted.
What this means
Installing the skill does not by itself verify that the local Morfeo Engine service is safe or the expected version.
Why it was flagged
The reviewed package does not include the Morfeo Engine implementation it controls, so this scan cannot validate the local backend behavior or provenance.
Skill content
Source: unknown; Homepage: none; No code files present — this is an instruction-only skill.
Recommendation
Review and pin the external Morfeo Engine repository/version separately before granting it tokens, PM2 control, or social-media publishing access.