Skillv1.0.0

ClawScan security

智能报告生成器 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousMar 16, 2026, 7:18 AM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill claims Feishu and database integration and requests python3, but it provides no credentials, no concrete access instructions, and has no code — these mismatches make its behavior unclear and warrant caution.
Guidance: This skill appears to be a high-level template for report generation but omits how it will access data or authenticate to Feishu/DBs. Before installing: 1) Ask the author how Feishu and database credentials are supplied and where they will be stored; 2) Confirm whether the platform provides connectors or if you must provide API keys/DB strings; 3) Disable or review any 'Auto Send' setting (Feishu) until credentials and destinations are verified; 4) If you must provide credentials, scope them minimally (e.g., a Feishu bot limited to a single doc) and avoid giving broad platform-level secrets; 5) Test in a sandbox or with throwaway data first; 6) Request a code implementation or clearer runtime instructions (how python3 is used) so you can review actual network calls and file access.

Review Dimensions

Purpose & Capability: concernThe description promises Feishu (飞书) integration, database queries, and automatic sending, but the skill declares no required environment variables, no config paths, and includes no code that would implement connectors. Requiring python3 while providing only templates and prose is disproportionate unless the platform supplies the actual connectors.
Instruction Scope: concernSKILL.md is high-level and instructs adding a TOOLS.md entry (Default Output: feishu_doc, Auto Send: true) but does not explain how to authenticate to Feishu, how database connections are supplied, or exactly which files/paths the agent should read. The vague, open-ended instructions could cause the agent to try available connectors or platform credentials without explicit guidance.
Install Mechanism: noteThere is no install spec or code to install; the only declared requirement is python3 (binary). For an instruction-only skill with no scripts, requiring python3 is unexpected but not directly dangerous — it is an incoherence that should be justified by the author.
Credentials: concernThe skill claims features that normally require secrets (Feishu API token, DB credentials, possibly file access) but declares no required env vars or config paths. That mismatch either omits necessary credential requirements or relies on implicit platform-level credentials — both situations are risky without clarification.
Persistence & Privilege: okalways is false and the skill does not request persistent or elevated privileges. Autonomous invocation is allowed (default) which is normal; there is no evidence the skill modifies other skills or global agent settings.