Trip Save

Save travel inspiration from YouTube videos, Instagram reels, blog URLs, tweets, or plain text like "want to visit Hampi". Extracts destination, auto-tags by...

Security Scan

The skill's requests and runtime instructions are coherent with its stated purpose (saving and tagging travel inspiration from URLs or text); it has no install steps, does not ask for credentials, and appears internally consistent.

Purpose & Capability

Name/description (save travel inspirations, extract destination and tags, store in memory) match the SKILL.md instructions. There are no unrelated required binaries, environment variables, or installs that would be unexpected for this functionality.

Instruction Scope

Instructions ask the agent to 'use web search or web fetch to get the page content' and then extract destination/tags and store a short record in memory. This is within purpose, but the fetch step means the agent will make network requests to user-provided URLs and could retrieve page content (which might include PII or private information). The skill does not instruct how to handle login-protected or paywalled content, or whether to store full page text versus only metadata.

Install Mechanism

Instruction-only skill with no install spec and no code files. Nothing is written to disk or downloaded during install, which is the lowest-risk install posture.

Credentials

The skill requests no environment variables, credentials, or config paths. That aligns with a bookmarking/saving feature that should not need external keys or secrets.

Persistence & Privilege

The skill stores saved items in agent memory (persistent by design). 'always' is false (no forced inclusion). Be aware that saved entries will persist according to the agent/platform memory retention and could be exposed in future agent responses or exports — the SKILL.md does not specify retention, redaction, or access controls.

Assessment

This skill appears to do only what it says: fetch public URLs or take plain text and save a small tagged record in agent memory. Before installing, consider: (1) Privacy — avoid sending URLs that contain sensitive information or private-account content because the agent will fetch page content and store metadata in memory. (2) Persistence — check your agent's memory retention and who can access it (saved items will persist and may appear in future conversations). (3) Protected content — the skill provides no mechanism for handling login-required or paywalled pages, so those saves may fail silently or save only the link. If you are comfortable with the agent making HTTP requests to the links you provide and storing simple records, this skill is proportionate to its purpose.

When to activate

• User forwards a URL (YouTube, Instagram, blog, tweet)
• User says "save this", "want to visit X", "adding X to my list", "bookmark this"

URL handling

1. Use web search or web fetch to get the page content
2. Extract: destination name, key details, vibe
3. Auto-tag and store in memory

Plain text saves

• "want to visit Hampi" → destination: Hampi
• Extract any vibe/budget hints from context

Auto-tagging

• Location: state/region (Himachal, Rajasthan, Goa, Kerala)
• Vibe: mountains, beaches, heritage, adventure, spiritual, nightlife, offbeat
• Budget: budget, mid-range, luxury
• Source: youtube, instagram, blog, text, voice

Memory storage

Remember each save as: "Saved destination: [name] | Tags: [tags] | Source: [url or text] | Date: [today]"

Recall

When user asks "show my saved places" or "what did I save":

• List all saved destinations from memory
• Support filtering: "show my beach saves"

Response

"✓ Saved: Kasol · mountains, Himachal, budget · from YouTube"

If destination unclear: "Saved the link. What place is this about?"

Rules

• Don't summarize the article/video — just save and confirm
• Don't give planning advice on save