Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
行李清单 Pro
v1.0.1场景化智能行李清单生成器。根据目的地、出行日期、出行目的和同行人，结合 FlyAI 实时搜索目的地信息，生成个性化分类行李清单。当用户提到"行李清单"、"带什么"、"打包"、"收拾行李"、"出行准备"时使用。
⭐ 0· 74·0 current·0 all-time
byhello_hang@hello-ahang
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill's purpose (scene-aware packing lists using FlyAI searches) matches the commands it plans to run (search-poi, keyword-search, etc.). However the SKILL.md instructs installing a global npm package at runtime and references different package names in places (e.g., @fly-ai/flyai-cli vs @anthropic-ai/flyai-cli), which is an inconsistency that should be clarified.
Instruction Scope
Instructions tell the agent to install or upgrade a global npm package at runtime and to run flyai CLI commands. Critically, it instructs running commands with NODE_TLS_REJECT_UNAUTHORIZED=0 to bypass SSL verification — this weakens transport security and is unnecessary for a packing-list generator. The skill also directs reading and writing a local profile file (~/.flyai/user-profile.md), which is reasonable for preferences but is sensitive filesystem access and should be explicit to users.
Install Mechanism
There is no formal install spec, but SKILL.md requires executing npm install -g @fly-ai/flyai-cli@latest at runtime. Global npm installs run arbitrary package code on the machine and require privileges; that increases risk compared to an instruction-only skill. The README/tools also references npx @anthropic-ai/flyai-cli@latest (different package namespace), adding ambiguity about which package will actually be installed/run.
Credentials
The skill declares no required credentials, which aligns with a simple packing-list tool. But it explicitly instructs setting NODE_TLS_REJECT_UNAUTHORIZED=0 for flyai CLI calls, which is a broad, unnecessary modification of TLS behavior and disproportionate to the stated task. The skill will also send user-provided destination/date info to an external FlyAI service via the CLI (expected), but the lack of declared endpoints or credential requirements means users can't easily audit where data is sent.
Persistence & Privilege
always is false and the skill does not request special system privileges. It does instruct creating/reading/writing a local file (~/.flyai/user-profile.md) for user preferences and to use platform memory APIs when available — this is a reasonable local persistence model but users should be informed that the skill will store profile data on disk.
What to consider before installing
This skill mostly matches its purpose, but there are several red flags you should consider before installing or running it: (1) The runtime tells you to run a global npm install; global packages execute code on your machine — prefer running packages with npx or inspect the package source first. (2) The SKILL.md inconsistently names the CLI package (@fly-ai/... vs @anthropic-ai/...) — ask the author which package and registry are authoritative. (3) It instructs setting NODE_TLS_REJECT_UNAUTHORIZED=0 to bypass TLS verification; never do this in general — it enables MITM attacks and is not justified for a packing-list generator. (4) The skill will read and write a local file (~/.flyai/user-profile.md) and will send your trip details to an external FlyAI service via the CLI; if that data is sensitive, avoid using the remote search or review the CLI's privacy/transport behavior first. If you want to use this skill safely: request clarification on the correct CLI package, inspect the package source (or run in an isolated sandbox/container), refuse the TLS bypass, and back up/delete the ~/.flyai directory as needed.Like a lobster shell, security has layers — review code before you run it.
latestvk973r3tf59bpkqq4q5rt3xjm05845ng6
License
MIT-0
Free to use, modify, and redistribute. No attribution required.