Travel Cog

AI travel planning and itinerary generation powered by CellCog. Vacation planning, travel research, flight planning, hotel recommendations, visa requirements...

Security Scan

The skill's instructions align with a travel-planning purpose but they depend on a third-party CellCog SDK and an API key that the skill metadata does not declare; that mismatch and the implicit data flow to an external service are concerning and worth verifying before install.

Purpose & Capability

The SKILL.md clearly depends on the 'cellcog' SDK and instructs setting CELLCOG_API_KEY, but the registry metadata declares no required env vars, no primary credential, and no required dependencies. That omission is incoherent: a travel planner that calls an external CellCog API legitimately needs an API key and a dependency, so the manifest should list them.

Instruction Scope

The runtime instructions stay within travel planning: examples, API usage, and output formats. However, they instruct the agent to call the CellCog API (client.create_chat), which will transmit user prompts, dates, preferences, and potentially personal travel details to an external service. The instructions do not tell the agent to read local files or unrelated system state.

Install Mechanism

There is no formal install spec (instruction-only), which reduces installer risk, but the SKILL.md recommends installing 'cellcog' via pip or using '/cellcog-setup'. That delegates execution to a third-party package; the package source (PyPI/GitHub) should be verified because the skill will rely on it.

Credentials

Although the skill metadata lists no required environment variables, the documentation explicitly instructs setting CELLCOG_API_KEY for authentication. This is a discrepancy: the skill will require a credential (an API key) to function, and that credential enables an external service to receive user data. The manifest should explicitly declare that requirement and the expected scope/permissions of that key.

Persistence & Privilege

No 'always: true' or unusual persistence is requested. The skill is user-invocable and can be invoked autonomously per platform defaults; nothing in the skill attempts to modify other skill configs or claim permanent system presence.

What to consider before installing

This skill appears to be a wrapper around the third-party CellCog service rather than a self-contained offline planner. Before installing or using it: (1) confirm the cellcog package source (PyPI/GitHub) and inspect its repository/release artifacts; (2) confirm CELLCOG_API_KEY is actually required and ask the publisher what scope/permissions that key has; (3) review CellCog's privacy policy to understand what user data (itinerary details, traveler names, dates, preferences) will be sent and stored; (4) if you need stronger assurances, request that the skill metadata be updated to declare the dependency and required env var explicitly; and (5) consider using a dedicated, limited-scope API key you can rotate if you proceed. If you are uncomfortable sending personal trip data to an external service or cannot verify the SDK/package provenance, do not install or enable the skill.

Travel Cog - AI Travel Planner Powered by CellCog

Real travel planning needs real research — not recycled blog listicles.

#1 on DeepResearch Bench (Apr 2026) applied to travel. CellCog researches current prices, visa requirements, weather patterns, local events, and hidden gems — then delivers complete itineraries as beautiful PDFs or interactive dashboards. Every recommendation grounded in fresh data, not outdated travel guides.

How to Use

For your first CellCog task in a session, read the cellcog skill for the full SDK reference — file handling, chat modes, timeouts, and more.

OpenClaw (fire-and-forget):

result = client.create_chat(
    prompt="[your task prompt]",
    notify_session_key="agent:main:main",
    task_label="my-task",
    chat_mode="agent",
)

All agents except OpenClaw (blocks until done):

from cellcog import CellCogClient
client = CellCogClient(agent_provider="openclaw|cursor|claude-code|codex|...")
result = client.create_chat(
    prompt="[your task prompt]",
    task_label="my-task",
    chat_mode="agent",
)
print(result["message"])

---

What You Can Plan

Complete Trip Itineraries

Day-by-day plans with logistics:

• City Trips: "Plan a 5-day trip to Tokyo for a first-time visitor who loves food and culture"
• Multi-City Tours: "Plan a 2-week Europe trip covering Barcelona, Rome, and Santorini"
• Business + Leisure: "Plan a week in Singapore — 3 days of meetings, 4 days exploring"
• Family Vacations: "Plan a 10-day family trip to Costa Rica with kids ages 6 and 10"
• Budget Travel: "Plan a backpacking route through Southeast Asia for 3 weeks under $2,000"

Example prompt:

"Plan a 7-day trip to Japan:

Travelers: Couple, late 20s, first time in Japan Dates: April 5-12, 2026 (cherry blossom season) Interests: Food (especially ramen and sushi), temples, photography, some nightlife Budget: Mid-range ($200-300/day for two) Base cities: Tokyo (3 nights), Kyoto (3 nights), Osaka (1 night)

Include: Day-by-day itinerary, restaurant recommendations, transport between cities (Shinkansen vs. bus), estimated costs, cherry blossom viewing spots, and tips for navigating without Japanese.

Deliver as a beautiful PDF I can reference on my phone."

Travel Research

Deep-dive research before you book:

• Destination Comparison: "Compare Bali vs. Thailand for a 2-week December honeymoon"
• Visa Research: "What visa requirements do US citizens need for a 30-day trip through Central America?"
• Safety & Health: "Research travel safety and health recommendations for Colombia in 2026"
• Seasonal Analysis: "When is the best time to visit Patagonia? Research weather, crowds, and costs by month"
• Local Events: "What festivals and events are happening in Portugal in September 2026?"

Logistics & Practical Info

The details that matter:

• Packing Lists: "Create a packing list for 2 weeks in Iceland in February"
• Budget Breakdowns: "Estimate daily costs for backpacking through South America"
• Transport Guides: "How to get around Japan — JR Pass vs. individual tickets for my itinerary"
• Accommodation Strategy: "Compare hotels vs. Airbnb vs. ryokans for my Japan trip"

---

Output Formats

Format	Best For
PDF Itinerary	Phone-friendly, printable, shareable with travel partners
Interactive HTML	Clickable maps, expandable days, budget tracker
Markdown	Integration into Notion, Obsidian, or other planning tools

Specify your preferred format. CellCog defaults to PDF when no format is specified.

---

Why CellCog for Travel?

Generic Travel AI	CellCog Travel Cog
Recycled top-10 lists	Fresh research on current prices and availability
One-size-fits-all	Tailored to your dates, budget, interests, and travel style
Text-only itineraries	Beautiful PDFs, interactive dashboards, or structured markdown
Surface-level tips	Deep research on visa, weather, events, hidden gems
Can't do logistics	Estimates costs, compares transport, suggests accommodations

---

Chat Mode for Travel

Scenario	Recommended Mode
Single trip itinerary	"agent"
Quick destination research	"agent"
Complex multi-country planning with deep research	"agent team"

Use "agent" for most travel planning.

---

Tips for Better Travel Plans

1. Be specific about dates: Prices, weather, and events vary dramatically by date.

2. Share your travel style: "Luxury resort", "budget backpacker", "boutique hotels" changes every recommendation.

3. Name your interests: "Food-focused", "adventure activities", "architecture and history", "beach and relaxation"

4. Set a budget: Even a rough range helps CellCog calibrate recommendations.

5. Mention constraints: "We don't drive", "vegetarian", "traveling with a toddler", "wheelchair accessible"

6. Ask for a PDF: CellCog's PDF itineraries are beautiful and phone-friendly — perfect for on-the-go reference.

---

If CellCog is not installed

Run /cellcog-setup (or /cellcog:cellcog-setup depending on your tool) to install and authenticate. OpenClaw users: Run clawhub install cellcog instead. Manual setup: pip install -U cellcog and set CELLCOG_API_KEY. See the cellcog skill for SDK reference.