Skillv1.0.0

ClawScan security

CIE 剑桥考试监考技能手册 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 22, 2026, 4:40 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: This is an instruction-only invigilation handbook whose content, requirements, and runtime instructions are internally consistent with its stated purpose (Cambridge exam invigilation guidance).
Guidance: This skill is coherent and appears to be a straight invigilation handbook, but note the source/homepage are unknown — before relying on it: (1) verify the wording and procedures against official Cambridge International documentation for the exact 2026 session, (2) confirm local/regional rules (metal-detection and confiscation procedures can be legally sensitive), and (3) prefer a published copy from Cambridge or your exam centre rather than an unsigned third-party file. If you will use this for official training, request provenance or an official reference to ensure no outdated or incorrect instructions are followed.

Purpose & Capability: okName and description describe invigilator training and the SKILL.md contains only procedural guidance for exam invigilation; there are no unrelated binaries, environment variables, or installs requested.
Instruction Scope: okSKILL.md stays within the scope of exam procedures, read-aloud scripts, checklists and handling of suspected malpractice. It does not instruct the agent to access local files, credentials, system state, or external endpoints beyond a single Cambridge reference URL.
Install Mechanism: okNo install spec and no code files (instruction-only). Nothing is written to disk or fetched at install time, which minimizes installation risk.
Credentials: okNo environment variables, credentials, or config paths are requested; required privileges are proportional to an instruction-only handbook.
Persistence & Privilege: okalways is false and model invocation is standard; the skill does not request persistent presence or modify system/other-skill settings.