Sogni Gen

SuspiciousAudited by ClawScan on May 10, 2026.

Overview

This skill mostly matches its Sogni image/video purpose, but a scan found a possible built-in Sogni API key, so review the source before trusting it with your account or media.

Before installing, inspect the reported SOGNI_API_KEY location in sogni-gen.mjs and use your own Sogni credentials only through the documented env vars or credentials file. Be mindful that prompts and selected reference media are sent to Sogni, and disable MCP downloads if you do not want result copies saved locally.

Findings (4)

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.

ConcernMedium Confidence
ASI03: Identity and Privilege Abuse
What this means

Requests could use or expose an unintended Sogni credential if the scan reflects an actual literal key.

Why it was flagged

The skill legitimately needs user-provided Sogni credentials, but a hardcoded API key in source would be an unintended credential boundary issue and could cause use of an unexpected account or expose a real key.

Skill content
Static scan at sogni-gen.mjs:1843: `SOGNI_API_KEY: [REDACTED]`
Recommendation

Review the source around the reported line, remove any literal API key, rotate any exposed key, and rely only on user-provided environment variables or the documented credentials file.

What this means

Personal photos, audio, or videos you choose as references may leave your device and generation may consume Sogni tokens.

Why it was flagged

The tool accepts local image, audio, and video files for editing, photobooth, and video workflows; this is central to the skill but means selected local media can be sent to Sogni.

Skill content
`node sogni-gen.mjs --photobooth --ref face.jpg ...` and `node sogni-gen.mjs --video --ref-audio song.mp3 ...`
Recommendation

Only provide media files you are comfortable uploading to Sogni, and confirm token/cost-sensitive requests before running them.

What this means

Installing the skill can fetch and run third-party npm packages needed by the Sogni CLI/MCP server.

Why it was flagged

Installation may execute npm dependency installation from the package manifest; this is normal for a Node-based runtime, but it introduces dependency provenance risk.

Skill content
command: "cd {{skillDir}} && cp skill-package.json package.json && npm i"
Recommendation

Install from the trusted publisher/source, prefer pinned or locked dependencies where possible, and review dependency changes on updates.

What this means

Prompts, seeds, URLs, or file references from prior renders may remain on disk and influence later generation requests.

Why it was flagged

The skill stores render metadata and can reuse the last image or seed in later runs; this persistence is disclosed and purpose-aligned.

Skill content
Last render metadata (read/write): `~/.config/sogni/last-render.json` ... `--last-image`
Recommendation

Review or delete the last-render file if prior outputs should not be reused, and avoid storing sensitive prompts or media references.