Travel Planning

Plan trips with itineraries, multi-city routing, budget optimization, family logistics, packing lists, and visa timelines.

This skill appears internally consistent: it is instruction-only, needs no external credentials, and stores trip data as local markdown files in ~/travel-planning/. Before installing or enabling it: 1) Confirm with the platform whether "MAIN memory" means a global agent memory (if you want to avoid cross-skill/global storage). 2) Be mindful what you share — the skill states it will only store passport expiry and reference numbers with consent and will not store full images; do not upload full document images. 3) Inspect and back up ~/travel-planning/ if you enable the skill; you can delete or move those files at any time. 4) If you prefer, disable autonomous invocation or limit its activation triggers so it only runs when you explicitly ask. 5) Because the source is 'unknown', if you need stronger assurance ask the publisher for more provenance (who maintains it, where source is hosted). Overall the behaviour matches the travel-planning purpose, but confirm the platform-level memory behavior if you want to avoid any global writes.

Type: OpenClaw Skill Name: travel-planning Version: 1.0.1 The OpenClaw AgentSkills bundle for 'travel-planning' is classified as benign. All files are markdown-based instructions and templates for the AI agent, with no executable code. The `SKILL.md` explicitly defines strict boundaries, stating the skill ONLY manages files within `~/travel-planning/` and NEVER reads files outside this directory, makes actual bookings, accesses email/calendar, or stores payment information. It also mandates user consent for storing limited sensitive data like passport expiry dates and explicitly forbids storing full document images. There is no evidence of data exfiltration, unauthorized execution, persistence mechanisms, obfuscation, or malicious prompt injection attempts against the agent.