Skillv1.0.0

ClawScan security

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

BenignMar 16, 2026, 6:59 AM

Verdict: Benign
Confidence: medium
Model: gpt-5-mini
Summary: The skill is internally consistent with its stated purpose (local fortune-telling using a bundled Python script) and requests no credentials or installs, but I reviewed a truncated copy of the script so final assurance is limited until the full file is inspected.
Guidance: This skill appears to be a self-contained entertainment fortune-telling tool that runs a local Python script and asks only for a birth date. Before installing or running it: (1) review the full scripts/liuyao_fortune.py file to confirm there are no hidden network calls, file writes, or subprocess spawns (the provided snippet contains only local computation but was truncated); (2) be aware you will be sharing personal data (birth date/time) with the agent — if you consider that sensitive, avoid entering exact birth times; (3) the SKILL.md uses an absolute path (/root/.openclaw/...), ensure the runtime path matches your environment; (4) if you have security concerns, run the script in a sandboxed environment or inspect its complete source before use.

Purpose & Capability: okName/description (六爻八卦算命) match the included assets: SKILL.md documents running a Python script and scripts/liuyao_fortune.py implements calendar/stem-branch, five-element scoring, and hexagram lookup. No unrelated cloud APIs, binaries, or credentials are requested.
Instruction Scope: okSKILL.md instructs the agent to run the bundled Python script with birthdate arguments and to present the script's textual report to the user. The instructions reference only the script path and the user's birthdate; they do not ask the agent to read arbitrary files, environment variables, or send data externally.
Install Mechanism: okNo install specification; the skill is instruction+script only. This is the lowest-risk model — nothing is downloaded or written at install time by the skill itself.
Credentials: okThe skill declares no required env vars, credentials, or config paths and the visible script uses only standard Python libs (sys, json, datetime). There is no apparent need for secrets or external service credentials.
Persistence & Privilege: okFlags are defaults (always: false, user-invocable: true). The skill does not request permanent presence or elevated privileges and does not modify other skills or global agent settings.