Skillv1.0.0

ClawScan security

Telegram语音消息技能包:基于实际踩坑经验的完整解决方案，帮助AI助手正确发送Telegram语音消息。解决WAV格式错误、缺少asVoice参数、TTS音频URL过期等常见问题。 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousMar 10, 2026, 9:52 PM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill's documentation and scripts match the stated purpose (TTS → convert → send Telegram voice messages) but the package metadata fails to declare the sensitive environment variables it needs and the runtime instructions/scripts perform network calls and credential use — this mismatch warrants caution.
Guidance: This package implements exactly what it claims (TTS → convert to OGG/libopus → send Telegram voice messages), but the metadata omitted the environment variables and credentials the scripts actually need. Before installing or running: 1) Inspect scripts (tts_generator.sh, telegram_sender.sh) to confirm the endpoints (Aliyun/OpenAI/Telegram) are what you expect; 2) Do not run with real production API keys — test in an isolated container/VM with throwaway credentials; 3) Ensure TELEGRAM_BOT_TOKEN and TTS API keys are not committed to VCS; 4) Verify log paths and the presence of any Python packages (e.g., json_log_formatter) before enabling logging to avoid unexpected installs; 5) If you need strict control, ask the author to update the skill metadata to declare required env vars (TELEGRAM_BOT_TOKEN, TELEGRAM_CHAT_ID, ALIYUN_TTS_API_KEY/OPENAI_API_KEY/etc.) so the platform can show clear permission requests. If you are uncomfortable with network calls or giving API keys, do not install/run the scripts.

Review Dimensions

Purpose & Capability: noteSKILL.md, README and the included scripts (tts_generator.sh, audio_converter.sh, telegram_sender.sh, etc.) all implement TTS generation, ffmpeg conversion and Telegram sending — this aligns with the skill name/description. However the registry metadata declares no required environment variables or primary credential while the instructions and scripts clearly require TELEGRAM_BOT_TOKEN, TELEGRAM_CHAT_ID and TTS API keys (ALIYUN_TTS_API_KEY, OPENAI_API_KEY, etc.). That metadata omission is an incoherence.
Instruction Scope: concernRuntime instructions and scripts explicitly download TTS audio URLs, call external TTS endpoints (dashscope.aliyuncs.com, api.openai.com) and perform network uploads to Telegram. They use environment variables for secrets and write/read temporary files under /tmp. This is expected for the stated task, but SKILL.md and scripts access credentials that are not declared in the package metadata, increasing surprise risk. There are no steps that request unrelated system secrets, but the skill does perform arbitrary network I/O and temporary file handling which the user should review.
Install Mechanism: okNo install spec is provided (instruction-only style), so nothing is automatically downloaded or executed on install. The package includes multiple shell scripts and a Python template that would run when invoked; absence of an automatic installer lowers supply-chain risk. Still, the included scripts will execute network calls when run.
Credentials: concernThe skill requires multiple sensitive environment variables in practice (TELEGRAM_BOT_TOKEN, TELEGRAM_CHAT_ID, ALIYUN_TTS_API_KEY, OPENAI_API_KEY, optional GOOGLE credentials) but the registry metadata lists none. Requiring multiple service credentials is reasonable for this functionality, but their absence from the declared requirements is a mismatch and a transparency problem. Also the templates/logging may write logs to /tmp and reference third-party Python logging formatters (json_log_formatter) which could cause additional package installs if used.
Persistence & Privilege: okThe skill does not request always:true and does not declare modifications to other skills or global agent settings. Scripts create and clean up temporary files; the skill does not appear to attempt persistent, privileged changes beyond writing to /tmp and optional log file paths configured in templates.