Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
多目的地航线组合比价助手
v1.0.0多目的地跨城比价航线优化助手，自动收集出行约束（出发地、假期时间窗、候选城市），一键比价筛出最优往返城市组合（更便宜+更少中转/更短飞行），输出Top3方案并给出可玩性建议。触发词：跨城比价、航线优化、机票比价、多城市飞、往返组合、机票怎么买最便宜、帮我搜机票、去程回程怎么选。
⭐ 0· 53·0 current·0 all-time
bylushengteng@purelanren
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill claims to perform flight searches and scoring — that aligns with the documented commands (flyai search-flight / keyword-search). However, the skill does not declare any required binaries or environment variables even though every core workflow runs the external 'flyai' CLI. Omitting 'flyai' as a required binary is an incoherence: either the platform already provides this binary (not documented here), or the skill will fail or attempt to install/execute an unexpected binary. This mismatch should be clarified.
Instruction Scope
The SKILL.md instructs the agent to: read/update Memory (search_memory / ask_user_question), run many flyai CLI commands in parallel, fetch external content (fetch_content), and create local HTML files (create_file). Those are reasonable for the stated purpose. Two problematic instructions: (1) all example commands prepend NODE_TLS_REJECT_UNAUTHORIZED=0, which disables TLS verification for outgoing HTTPS requests — this weakens security and can enable MITM; (2) the skill expects to surface booking URLs (jumpUrl) which are external links the user may click — the skill does not include any verification or sanitization guidance for these links. Also, the runbook exercises broad discretion (parallel searches, alias fallback, date expansion) but that's within purpose.
Install Mechanism
This is instruction-only (no install spec), which lowers risk. However, it depends heavily on an external CLI ('flyai') and platform tools (search_memory, ask_user_question, fetch_content, create_file). The absence of an install spec is acceptable if the execution environment already provides these tools — but that assumption is not stated in the metadata, which is a documentation/incoherence issue rather than a direct install risk.
Credentials
The skill declares no required environment variables, but the instructions rely on setting NODE_TLS_REJECT_UNAUTHORIZED=0 in command invocations — an environment variable used to disable SSL verification. That environment usage is not declared and is security-reducing. Aside from that, the skill does not request credentials. The skill does read and update user Memory (user profile, preferences) which is expected for personalization but has privacy implications; the skill does not describe retention or opt-out behavior.
Persistence & Privilege
The skill is not 'always: true' and does not request elevated privileges or system-wide configuration changes. It uses Memory to store user preferences (normal for personalization) and creates files in the user's working directory. Autonomous invocation is allowed by default but is not combined with high privilege or persistent self-enablement. No concerning persistence behavior observed.
What to consider before installing
Before installing, confirm these points: (1) Verify whether your agent platform already provides the 'flyai' CLI and the platform tools (search_memory, ask_user_question, fetch_content, create_file). If not, the skill will fail or will require installing unknown binaries. (2) Ask the skill author why every command disables TLS verification (NODE_TLS_REJECT_UNAUTHORIZED=0). Disabling TLS verification is unsafe — do not run commands with this environment variable set unless you understand and accept the risk. (3) Understand privacy: the skill reads and updates Memory (travel origin, preferences, companions). If you care about sensitive data, ask how long that memory is stored and how to opt out. (4) The skill surfaces external booking links (jumpUrl). Treat these like any external link — verify destination before clicking. (5) If you decide to proceed, run the skill in a sandboxed environment first (or request a version that does not disable TLS) and verify that the 'flyai' CLI used is the official / trusted implementation. If the author can confirm that the runtime environment already provides flyai and can remove or justify the TLS-disable pattern, the incoherences would be resolved and my confidence would increase.Like a lobster shell, security has layers — review code before you run it.
latestvk973526nhewrzcdcs82vredpmh840n9j
License
MIT-0
Free to use, modify, and redistribute. No attribution required.