Al Music Generation

PassAudited by ClawScan on May 1, 2026.

Overview

The skill is a coherent ShortAPI music-generation integration that requires a ShortAPI key and uses bounded background polling, with no artifact-backed evidence of deceptive or destructive behavior.

Before installing, confirm you are comfortable sending prompts and generation requests to ShortAPI, provide only a ShortAPI key, avoid putting secrets in prompts or callback URLs, and ensure the agent limits polling to user-started jobs for no more than 5 minutes.

Findings (3)

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

Using the skill lets the agent make ShortAPI job requests with the user's ShortAPI key.

Why it was flagged

The skill requires a ShortAPI credential and uses it to authenticate API calls. This is expected for the stated integration, but it is still account authority the user should protect.

Skill content
"requires": { "env": ["SHORTAPI_KEY"] } ... "Authorization: Bearer $SHORTAPI_KEY"
Recommendation

Use only a ShortAPI-specific key, keep it in the environment rather than chat, and rotate it if exposed.

What this means

A changed or overly broad remote model document could influence how the agent builds ShortAPI requests.

Why it was flagged

The agent relies on a runtime-fetched Markdown document to construct API arguments. That is purpose-aligned, but retrieved content should be limited to schema interpretation and not treated as authority for unrelated instructions.

Skill content
You **MUST** first fetch the detailed skill document ... The document returned in Step 1 is the sole source of truth for the model's input schema.
Recommendation

Treat fetched model documents as parameter schemas only; ignore unrelated instructions, credential requests, or tool-use directions in those documents.

What this means

After starting a generation job, the agent may keep querying ShortAPI for status for up to 5 minutes without further prompts.

Why it was flagged

The skill directs autonomous background polling after a job starts. The behavior is disclosed and bounded, but users should understand the agent may continue making status requests for a short period.

Skill content
The Agent should poll the status endpoint for a maximum of **5 minutes** per job ... You **MUST** continue to poll the status endpoint in the background silently.
Recommendation

Allow polling only for user-started jobs and ensure the 5-minute limit and current-conversation-only state handling are enforced.