← Back to Skills Marketplace
senseaudio-game-npc-director
by
Wu Ruixiao
· GitHub ↗
· v1.0.1
· MIT-0
248
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install senseaudio-game-npc-director
Description
Use when a game, interactive story, or virtual world needs reusable NPC voice behavior, including fixed voice identity, catchphrases, relationship-aware dial...
Usage Guidance
Before installing or enabling this skill, verify the following: (1) Confirm which environment variables and config files you must provide — the code expects SENSEAUDIO_API_KEY and/or SENSEAUDIO_PLATFORM_TOKEN, an ASR model env, and a Feishu app_id/app_secret stored in a workspace config, but the registry metadata lists none. (2) Ask the publisher where the _shared helper modules come from (senseaudio_env, audioclaw_paths, senseaudio_api_guard) and the audioclaw-skills-voice-reply feishu helper; these are not included and will be imported from the host environment. (3) Understand that the skill will send audio to external endpoints (api.senseaudio.cn, platform.senseaudio.cn, open.feishu.cn) and will upload/transcode audio — do not supply sensitive audio or credentials unless you trust the endpoints and code. (4) If you cannot confirm the provenance of the missing helpers and the required credentials, run the skill only in an isolated environment or decline installation. If you decide to proceed, ask the maintainer to update the registry metadata to declare required env vars and config paths and to include or document any dependent helper packages.
Capability Analysis
Type: OpenClaw Skill
Name: senseaudio-game-npc-director
Version: 1.0.1
The skill bundle is a comprehensive toolkit for creating game NPC voice interactions using SenseAudio ASR/TTS and Feishu for delivery. The scripts (e.g., batch_tts_scene.py, senseaudio_asr.py) perform legitimate network requests to official API endpoints (api.senseaudio.cn) and handle local file operations for audio processing. While the orchestrator (run_player_voice_npc_pipeline.py) uses subprocesses and accesses a specific credential file (~/.audioclaw/workspace/state/senseaudio_credentials.json), these actions are transparently documented and strictly aligned with the functional requirements of the NPC director role.
Capability Assessment
Purpose & Capability
The scripts and SKILL.md implement NPC voice generation, ASR, TTS, and Feishu delivery which matches the skill description. However, the registry metadata lists no required env vars or config paths even though the implementation clearly expects API keys, platform tokens, and host helper modules.
Instruction Scope
Runtime instructions and scripts call external services (SenseAudio ASR/TTS endpoints, Feishu open API), transcode and upload audio, and load helper modules from parent _shared directories and another skill (audioclaw-skills-voice-reply). They also read workspace/config files via audioclaw_paths. The SKILL.md and registry do not disclose these file/config dependencies or the exact credentials used, giving the agent access to external networks and workspace configs beyond what's declared.
Install Mechanism
No install spec is provided (instruction + scripts only), so nothing arbitrary is downloaded at install time. This lowers install-time risk. The code will run at runtime and import helper modules from the host environment.
Credentials
The code uses SENSEAUDIO_API_KEY, SENSEAUDIO_PLATFORM_TOKEN, SENSEAUDIO_ASR_MODEL (and an api-key-env argument), and expects Feishu app credentials/config (app_id/app_secret) via a workspace config — none of these were declared in the registry metadata. Requiring platform API keys and tenant credentials is normal for ASR/TTS and Feishu features, but the omission in metadata is an incoherence that could lead to surprise credential usage or accidental credential leakage.
Persistence & Privilege
The skill is not marked always:true and does not modify other skills' configurations. It does import shared modules from parent directories, but there is no evidence it attempts to persist itself or alter system-wide agent settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install senseaudio-game-npc-director - After installation, invoke the skill by name or use
/senseaudio-game-npc-director - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- Major terminology update: "SenseAudio" renamed to "AudioClaw" throughout all user-facing documentation and skill descriptions.
- `scripts/run_player_voice_npc_pipeline.py` now supports both `--input-audio` and `--input-text`, enabling seamless NPC reply generation from either player speech or text.
- NPC voice reply mode now defaults to voice/audio unless the user explicitly requests text-only output.
- Detailed session management and reply mode logic clarified, ensuring NPC dialogue mode stays "sticky" for ongoing turns.
- API key lookup and injection updated: TTS scripts now default to using `SENSEAUDIO_API_KEY`, with improved logic for real API key substitution during voice synthesis. ASR script behavior remains unchanged.
v1.0.0
Initial release of senseaudio-game-npc-director.
- Enables reusable, fixed-identity NPC voice generation for games and interactive worlds using SenseAudio.
- Supports relationship-aware dialogue, catchphrases, event-based lines, world narration, and ASR-driven NPC reactions to player speech.
- Provides scripts for scene building, player voice intake, TTS synthesis, and audio delivery to platforms like Feishu.
- Maintains session state for each NPC, supporting dynamic player interaction and voice switching.
- Focuses on low-cost, high-immersion NPC voice assets, suitable for both static assets and fully interactive voice experiences.
Metadata
Frequently Asked Questions
What is senseaudio-game-npc-director?
Use when a game, interactive story, or virtual world needs reusable NPC voice behavior, including fixed voice identity, catchphrases, relationship-aware dial... It is an AI Agent Skill for Claude Code / OpenClaw, with 248 downloads so far.
How do I install senseaudio-game-npc-director?
Run "/install senseaudio-game-npc-director" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is senseaudio-game-npc-director free?
Yes, senseaudio-game-npc-director is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does senseaudio-game-npc-director support?
senseaudio-game-npc-director is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created senseaudio-game-npc-director?
It is built and maintained by Wu Ruixiao (@kikidouloveme79); the current version is v1.0.1.
More Skills