travel-agent-skills
llms.txt4 starsChanged 23 days agoMITView on GitHub
Agent skills for live flight and hotel pricing with Google's low|typical|high verdict FlightPowers Travel Agent Skills — eight prompt-driven workflows for Claude, Cursor, and any agent runtime that reads a skills directory. Built on two independent real-time APIs: Google Flights Live API and Booking Live API, both available on RapidAPI. Not affiliated with, endorsed by, or sponsored by Google or Booking.com. Gives your agent the data and the workflow to answer real travel questions: Every fare carries price_insights_low, price_insights_high, and…
- Reads credentials
In this repository8 other files it gives its agents.
- cheapest-datesSkill
- destination-compareSkill
- fare-watchSkill
- hotel-searchSkill
- rate-parity-monitorSkill
- travel-briefSkill
- travel-data-apiSkill
- trip-plannerSkill
The file
We split this llms.txt into the rules it gives an agent
7 rules. About 0 are common, 5 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Distinctive: line 9
Not affiliated with, endorsed by, or sponsored by Google or Booking.com.
- Very specific: line 22
Every fare carries price_insights_low, price_insights_high, and price_range_in_relation_to_other_periods — Google's own historical band for …
- Distinctive: line 78
Prefer MCP tools when available.
- Very specific: line 80
Check X-Search-Status (REST) or the tool's outcome note (MCP) before reporting a route as empty
- Distinctive: line 81
Never cache or reuse a fare from earlier in the conversation — always fetch fresh
- Distinctive: line 83
Never print, log, or commit an API key
- Distinctive: line 84
No invented numbers — there are no uptime, latency, user-count or revenue claims anywhere in this repo
≈ “Common” is an estimate: at least 1 in 5 of the 3,727 llms.txt files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
# Travel Agent Skills > Agent skills for live flight and hotel pricing with Google's low|typical|high verdict ## Product FlightPowers Travel Agent Skills — eight prompt-driven workflows for Claude, Cursor, and any agent runtime that reads a skills directory. Built on two independent real-time APIs: Google Flights Live API and Booking Live API, both available on RapidAPI. Not affiliated with, endorsed by, or sponsored by Google or Booking.com. ## What it does Gives your agent the data and the workflow to answer real travel questions: - **Cheapest dates**: "When should I fly Berlin to Paris in late June?" → one call searches five dates, ranked by price, each with Google's `low | typical | high` verdict - **Destination compare**: "Somewhere warm under $400 in October from JFK" → fan out to multiple cities in one search - **Trip planner**: "7 nights in Lisbon under $2000 for two, hotel needs free cancellation" → aligned flights + hotels, totalled per person - **Fare watch**: Track one route over time, alert only on a meaningful drop - **Hotel search**: Real-time Booking.com availability with 24 filters - **Rate parity monitor**: Price the same hotel as a resident of several countries would see it Every fare carries `price_insights_low`, `price_insights_high`, and `price_range_in_relation_to_other_periods` — Google's own historical band for that route and period. That is the difference between an agent that says "$49" and one that says "$49 is low for this route — book it, don't wait." ## Install (60 seconds) ### MCP servers (ad-free; sign in with Google, no key to start) ```bash claude mcp add --transport http flights https://flights.flightpowers.com/mcp claude mcp add --transport http hotels https://hotels.flightpowers.com/mcp ``` Four tools (one-way flights, round-trip, hotel search, hotel by name). Signed in with Google, the first 10 searches a day are free with no key (one search = one date x destination combination, counted across both servers). After that, paste your RapidAPI key once on the /connect page (https://flights.flightpowers.com/connect), or send it as a header. ### With your own RapidAPI key as a header ```bash # Get a key at https://rapidapi.com/mtnrabi/api/google-flights-live-api # Same key for hotels at https://rapidapi.com/mtnrabi/api/booking-live-api claude mcp add --transport http flights https://flights.flightpowers.com/mcp \ --header "x-rapidapi-key: YOUR_RAPIDAPI_KEY" claude mcp add --transport http hotels https://hotels.flightpowers.com/mcp \ --header "x-rapidapi-key: YOUR_RAPIDAPI_KEY" ``` ### Skills (prompt workflows) ```bash # Claude Code plugin claude plugin marketplace add mtnrabi/travel-agent-skills claude plugin install travel-agent-skills@travel-agent-skills # Or any agent with a skills directory git clone https://github.com/mtnrabi/travel-agent-skills.git cp -r travel-agent-skills/skills/* ~/.claude/skills/ ``` ## APIs and MCPs | Product | What | Where | |---|---|---| | Google Flights Live API | Real-time airfares with Google's price insights | https://rapidapi.com/mtnrabi/api/google-flights-live-api | | Booking Live API | Real-time hotel availability and nightly rates | https://rapidapi.com/mtnrabi/api/booking-live-api | | Flights MCP (ad-free) | Hosted MCP server: 10 free searches a day after Google sign-in, then your RapidAPI key | https://flights.flightpowers.com/mcp | | Hotels MCP (ad-free) | Hosted MCP server: same sign-in allowance, then your RapidAPI key | https://hotels.flightpowers.com/mcp | ## Documentation - Full README: https://github.com/mtnrabi/travel-agent-skills/blob/main/README.md - Skills reference: https://github.com/mtnrabi/travel-agent-skills/tree/main/skills - API reference: https://github.com/mtnrabi/travel-agent-skills/blob/main/skills/travel-data-api/SKILL.md - Homepage: https://flightpowers.com ## Design rules 1. Prefer MCP tools when available; fall back to REST otherwise 2. Say what a call costs before making it — every billed request is the user's money 3. Check `X-Search-Status` (REST) or the tool's outcome note (MCP) before reporting a route as empty 4. Never cache or reuse a fare from earlier in the conversation — always fetch fresh 5. Lead with the answer, then the evidence 6. Never print, log, or commit an API key 7. No invented numbers — there are no uptime, latency, user-count or revenue claims anywhere in this repo ## License MIT — see https://github.com/mtnrabi/travel-agent-skills/blob/main/LICENSE
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

