Skillv1.0.0

ClawScan security

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

BenignApr 18, 2026, 8:21 AM

Verdict: Benign
Confidence: high
Model: gpt-5-mini
Summary: This is an instruction-only skill that asks the agent to provide public information about Zurich and does not request credentials, install anything, or access system files—its declared purpose matches its runtime instructions.
Guidance: This skill is low-risk and coherent: it simply guides the agent to provide public information about Zurich. Before installing, note that (1) it relies on web/public sources — verify facts against official Zurich publications or reputable news outlets, (2) the agent may produce outdated or hallucinated details, so cross-check critical data, and (3) the skill does not request credentials or access to local files; if the skill later asks for any secret or system access, treat that as suspicious and revoke it.

Purpose & Capability: okThe name/description (providing Zurich's history, business scope, global presence and news) matches the SKILL.md content. No unrelated binaries, env vars, or capabilities are requested.
Instruction Scope: okSKILL.md only defines information modules and simple triggers (search/consult). It does not instruct reading local files, accessing credentials, or sending data to unexpected endpoints. It implicitly relies on web/public sources for news, which is consistent with the stated purpose.
Install Mechanism: okNo install specification or code files — instruction-only skills are lowest-risk because nothing is written to disk or executed locally by the skill itself.
Credentials: okNo environment variables, credentials, or config paths are requested; the skill's needs are minimal and appropriate for a public-information task.
Persistence & Privilege: okalways:false and default agent invocation are used. The skill does not request persistent or system-wide changes and does not modify other skills' configs.