Skillv1.0.0

ClawScan security

Waiting Tips · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignMar 7, 2026, 11:57 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's code, files, and runtime instructions are consistent with its description: it loads local bilingual tip text files and exposes a tool to return/display them while the agent is thinking.
Guidance: This skill appears to do what it says: it reads local tips files and returns formatted tip text for display. Things to consider before installing: (1) The adapters shown are examples — only Telegram, Feishu, and WhatsApp adapters are provided in code; other platforms mentioned in the docs are not implemented. (2) Running the example adapters requires platform credentials (Telegram Bot token, Feishu APP_ID/APP_SECRET, or WhatsApp auth) and the WhatsApp example writes auth state to disk — only provide credentials you trust and run examples in a secure environment. (3) The skill will send a message before each AI response if you wire it into the default flow, which could be noisy in group chats; review/adjust automatic usage if that matters. Otherwise the files contain no network callbacks or hidden endpoints and no unexplained requests for secrets.

Review Dimensions

Purpose & Capability: okThe name/description (waiting tips while OpenClaw thinks) matches the implementation: core modules load tips/*.txt and return random bilingual tips in several display styles. Required capabilities are minimal and proportional — no unexpected credentials, binaries, or external services are required by the core skill.
Instruction Scope: noteSKILL.md instructs sending a tip before the AI response and describes adding tips by editing tips/*.txt — that matches the code. Small inconsistencies: README and SKILL.md claim support for many platforms (Discord, Slack, iMessage) but the provided adapters are only for Telegram, Feishu, and WhatsApp; those other platforms are not present in the code. Examples show integrations that require platform credentials (e.g., BOT_TOKEN, APP_ID/APP_SECRET, WhatsApp auth files) — those are example usages, not required by the skill itself, but users should be aware supplying those credentials is necessary to run the adapters.
Install Mechanism: okThere is no automatic install script fetched from the network; package.json and a local build.sh/test are included. README suggests cloning from a GitHub repo or local install — both are standard and traceable. No downloads from arbitrary URLs or archive extraction are present in the manifest.
Credentials: okThe skill does not declare or require any environment variables or secrets. Example adapters (not required to use the core skill) demonstrate typical platform credentials (Telegram bot token, Feishu app id/secret, WhatsApp auth) — these are expected for those adapters but are optional and not requested by the core plugin itself.
Persistence & Privilege: okThe skill does not request 'always: true' and is user-invocable. It registers an optional tool (optional: true) and does not modify other skills or system-wide settings. Example adapters may write auth files (WhatsApp example), but that's part of running those adapters and not the core skill requesting elevated privileges.