提供有声书创作与音频能力（ABS 读写、音效/音频检索、二创、音色推荐、章节角色分析等），通过 HTTP Streamable MCP 调用。
Security Analysis
high confidenceThe skill's declared purpose (audio/audiobook MCP calls) matches its instructions and required token; it is an instruction-only integration that sends user-provided content to aigc.ximalaya.com and requires a single bearer token (AUDIOBOOKLM_TOKEN).
Name/description, declared endpoint (https://aigc.ximalaya.com/audiobooklm/mcp), and required env var (AUDIOBOOKLM_TOKEN) are coherent: a network API token is expected for an HTTP MCP-based audio service.
SKILL.md contains explicit MCP request flow, headers, tool list, parameter schemas, and failure-handling rules. All runtime instructions relate to calling the remote MCP endpoint. Important: the skill explicitly sends user text/audio URLs/structured data to the third-party domain, so user content will leave the local environment.
No install spec or code files — instruction-only skill. This minimizes local-code risk (nothing downloaded or written), but runtime network calls still occur.
Only a single required credential (AUDIOBOOKLM_TOKEN) is requested and declared as the primary credential, which is proportionate. Caveat: that token is a bearer token sent to the remote service; the skill strongly warns about not sending sensitive/unauthorized data and recommends using low-privilege/test tokens and rotation.
always is false, no system config paths requested, and the skill does not request persistent system-level privileges. Autonomous invocation is allowed (platform default) but not elevated by this skill.
Guidance
This skill is internally consistent with its stated purpose. Before installing: only proceed if you trust https://aigc.ximalaya.com; provide a low-privilege or test AUDIOBOOKLM_TOKEN and rotate/revoke it if compromised; do not submit secrets or sensitive/confidential data (the skill will send content to the remote service); review the provider's privacy and terms to understand how uploaded audio/text is used; be aware the agent may call this skill autonomously when relevant — limit token scope or disable the skill if you want to prevent automatic remote calls.
Latest Release
v0.0.5
- Removed legacy configuration file audiobooklm.yaml for streamlined config management. - SKILL.md metadata updated: added a "clawdbot" section with an explicit "primaryEnv" definition. - No user-facing feature or API changes; this release is a minor maintenance update for compatibility and metadata precision.
Popular Skills
Published by @audiobooklm on ClawHub
