Kie Audio Generator

v1.0.0

Generate music and audio via Kie.ai's Suno gateway (V3.5 through V5.5). Use for background tracks, instrumental beds, full songs with vocals, or extending ex...

0· 35·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for benhuebner01/kie-audio.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Kie Audio Generator" (benhuebner01/kie-audio) from ClawHub.
Skill page: https://clawhub.ai/benhuebner01/kie-audio
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Required env vars: KIE_API_KEY
Required binaries: python3
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install kie-audio

ClawHub CLI

Package manager switcher

npx clawhub@latest install kie-audio
Security Scan
Capability signals
CryptoRequires sensitive credentials
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description (Suno/Kie audio generation) aligns with required binary (python3), required env (KIE_API_KEY), and the included client + generator scripts. The skill includes code to create/poll Suno generation tasks and download resulting audio, which is expected for this purpose. Note: the package includes executable Python scripts (not just prose), which is reasonable but worth noting since registry metadata listed no install spec.
Instruction Scope
SKILL.md instructs the agent to run the included Python scripts and describes the CLI options (generate, extend, webhook mode). The scripts perform HTTP requests to api.kie.ai, poll for task state, and download audio/cover URLs returned by the service — all within the scope of audio generation. The scripts also provide a webhook mode that listens on a local port for callbacks (expected). They read ~/.openclaw/openclaw.json to auto-load env vars (explicitly documented in SKILL.md), which expands their scope to any keys present in that file.
Install Mechanism
No install spec; the code is bundled and uses only Python stdlib (urllib/httpserver/etc.), matching its 'zero-install' claim in the client docstring. There is no download-from-unknown-URL behavior at install time; runtime network calls are to the declared API host.
!
Credentials
The skill legitimately requires KIE_API_KEY (primary credential) and optionally KIE_WEBHOOK_HMAC_KEY for webhook verification. However, the client auto-loads every key under ~/.openclaw/openclaw.json → globalEnv into the process environment if not already set. That means secrets not declared by the skill (other API keys, tokens) could become available to the skill at runtime. This behavior is documented in the code/SKILL.md but is broader than the single declared env var and could expose unrelated credentials if the user has other secrets stored there.
Persistence & Privilege
Skill does not request always:true and does not modify other skills or system-wide settings. It reads a user config file but does not write global configuration or persist itself beyond the included files. Autonomous invocation is allowed (platform default) but not combined with other high-risk privileges.
Assessment
This skill appears to do what it claims: it uses KIE_API_KEY to talk to https://api.kie.ai, create/poll generation tasks, and download generated audio. Before installing or running: 1) Inspect your ~/.openclaw/openclaw.json — the skill will auto-load all keys under its globalEnv into the process environment if they are not already set. Remove or move any unrelated secrets you don't want the skill to see. 2) Only provide a KIE_API_KEY you trust and revoke it if you later suspect misuse. 3) If you use webhook mode the script will open a local HTTP listener on the specified port — ensure that port is safe to use in your environment. 4) The script will download audio/cover URLs returned by the service (including from arbitrary URLs present in API responses); review output files and consider running in an isolated workspace if you are concerned about untrusted content. 5) If you need higher assurance, review the full included scripts (generate.py and kie_client.py) yourself before running.

Like a lobster shell, security has layers — review code before you run it.

Runtime requirements

🎵 Clawdis
Binspython3
EnvKIE_API_KEY
Primary envKIE_API_KEY
latestvk97dx3qwp1p9e0p09p0e6ctefs85p7sk
35downloads
0stars
1versions
Updated 8h ago
v1.0.0
MIT-0

kie_audio — Suno music generation via Kie.ai

Generate music and audio via Kie.ai's Suno gateway. Handles Suno's multi-stage state machine (PENDING → TEXT_SUCCESS → FIRST_SUCCESS → SUCCESS) and downloads all generated tracks (Suno often returns variations) when done.

When to use

  • User wants a background track, intro, outro, instrumental bed.
  • User wants a song with vocals from a lyric or story prompt.
  • User explicitly says Suno, V5, V4.5+, etc.
  • User wants to extend an existing audio clip.

Models

--modelNotes
V3_5Oldest, cheapest
V4
V4_5
V4_5PLUS
V5
V5_5 (default)Latest, highest quality

How to invoke

Use the exec tool. {baseDir} = this skill's folder.

Instrumental lofi loop

exec:
  command: python3 {baseDir}/scripts/generate.py --model V5_5 --instrumental --prompt "lofi hip hop, rainy afternoon, mellow piano, soft vinyl crackle, 30 seconds" --out ./out
  yieldMs: 1000

Full song with vocals (custom mode)

exec:
  command: python3 {baseDir}/scripts/generate.py --model V5 --custom-mode --prompt "upbeat indie pop, chorus about summer road trips, male vocal" --out ./out```

### Extend existing audio

exec: command: python3 {baseDir}/scripts/generate.py --extend https://example.com/original.mp3 --prompt "continue with a building bridge then a final chorus" --model V5_5 --out ./out```

Full CLI reference

python3 {baseDir}/scripts/generate.py \
    --prompt "..." \
    [--model V3_5|V4|V4_5|V4_5PLUS|V5|V5_5] \
    [--instrumental] [--custom-mode] \
    [--extend AUDIO_URL] \
    [--out ./out] \
    [--callback-url URL] [--callback-port 8787] [--no-wait] [--timeout 900]

Output

The script prints JSON to stdout:

{
  "taskId": "abc123...",
  "tracks": [
    {"audio": "./out/abc123_1.mp3", "duration": 125, "cover": "./out/abc123_1_cover.jpg"},
    {"audio": "./out/abc123_2.mp3", "duration": 130, "cover": "./out/abc123_2_cover.jpg"}
  ]
}

Important

  • Suno usually generates multiple tracks per request — the script downloads all of them plus their cover images.
  • State transitions are logged to stderr so you can see progress during polling.
  • Failure codes to watch: 400 (possible copyright), 413 (audio conflict), 501 (generation failed), 531 (server error — Kie auto-refunds credits).
  • All tracks expire at 14 days, download immediately.

Environment

Both are auto-loaded from ~/.openclaw/openclaw.jsonglobalEnv by the bundled script. No need to pass env: into exec.

Comments

Loading comments...