Flightaware logo

Flightaware

Community
win4r
flightaware

Use for questions about a specific flight’s departure or arrival time, including “when’s my flight?” and confirmation of remembered times, plus flight status, delays, and cancellations. Verify the exact dated flight before answering; memory identifies the itinerary but does not verify its current schedule. Use FlightAware to monitor operational changes for an upcoming booked flight when the user directly asks for ongoing monitoring.

Overview

Publisherwin4r
RepositoryMuseAI-Skills
Skill nameflightaware
Stars
382
Forks
116
Bundled files
3
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by win4r on GitHub. Read the source before you install it.

Installation

Install the Flightaware AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/win4r/MuseAI-Skills/tree/main/opt/hatch/skills/flightaware
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/win4r/MuseAI-Skills.git /tmp/MuseAI-Skills
mkdir -p .claude/skills
cp -r /tmp/MuseAI-Skills/opt/hatch/skills/flightaware .claude/skills/flightaware
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Flightaware in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Flightaware on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Flightaware is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

FlightAware

Connecting

There is nothing for the user to connect and no key to enter. Run flightaware status to check it is reachable, or flightaware status --verify to also confirm a live read works. If status comes back unavailable, tell the user FlightAware is not reachable from this device right now and try again later. Never ask the user for a FlightAware API key.

Common flows

Track a flight

Run flight with the airline flight number to get its current status, then use the returned fa_flight_id with position for where it is now or track for its recent path. For a picture, run map with --save <path> to write a static image of the flight.

Monitor a tracked trip

Use this flow when the user directly asks for ongoing monitoring of an upcoming booked flight. Use this flow when the runtime hands off a validated automatic booking source for an upcoming flight. Treat either trigger as authorization to create the watch after the required flight facts are present. Do not ask whether to start monitoring after the runtime handoff. Do not require a connected email account. Ask the user only for flight facts that the requested monitoring needs and that the current context does not supply. Resolve each exact dated leg before scheduling a job.

Search user goals and Tracking items for the trip. Use a matching user goal as the canonical owner. Otherwise, reuse a matching active Tracking item. If neither exists, create one Tracking item. Resolve the canonical owner before scheduling a job. Do not create a FlightAware-specific database or a duplicate Tracking item.

Store the minimum operational state in ~/workspace/goals/<goal-slug>/hidden_files/travel/flight.json. Use ~/workspace/goals/<goal-slug>/hidden_files/travel/flight.json only as goal-owned job state. Do not create a separate travel database.

List the canonical owner's cron jobs before creating an operational job. Reuse an equivalent goal-owned operational job when one exists. If no equivalent job exists, add one operational job for the trip. Use flight-status-<goal-slug> as its identifier. Set its owner to goal:<goal-slug>. Schedule the operational job at creation with the actual cron schedule for the next active leg's current cadence band. Use a daily cron schedule while the next active leg departs more than 48 hours from now. Use an hourly cron schedule from 48 hours before departure until 6 hours before departure. Use a 10 to 15 minute cron schedule from 6 hours before departure through landing. Do not create or retain a high-frequency cron schedule that relies on ~/workspace/goals/<goal-slug>/hidden_files/travel/flight.json to skip source calls. Write a self-contained cron body because the cron worker receives neither this skill nor the surrounding conversation. Include the Tracking goal, ~/workspace/goals/<goal-slug>/hidden_files/travel/flight.json, exact leg identity, and booked baseline. Include the deduplication rule, failure policy, cadence transitions, and stop conditions.

At the start of each operational run, compare the existing cron schedule with the next active leg's current cadence band. Update the same cron job before any FlightAware source call when the existing cron schedule differs from the current cadence band. Repair an existing high-frequency cron schedule before any FlightAware source call when the next active leg is outside the high-frequency band. Keep the current cadence until the active leg lands or a connection risk remains unresolved. Advance the same job to the next active leg, then stop it after the final landing.

Compare strategic schedule changes with the booked baseline. Compare operational changes with the last notified facts. Apply the cancellation corroboration and status interpretation rules in the Rules section. For a material movement, flight or carrier change, confirmed cancellation, diversion, terminal change, qualifying gate change, qualifying equipment change, qualifying delay, or credible connection risk, create one stable event fingerprint. Treat the first delay estimate at 30 minutes or more as a qualifying delay. Treat a later delay-only change as a qualifying delay only when it crosses from 30 minutes or more to less than 30 minutes, enters a worse band among 30 to 59 minutes, 60 to 119 minutes, and 120 minutes or more, or differs by 60 minutes or more from the last notified delay estimate. Do not include the exact delayed minute value in a delay-only fingerprint. Treat terminal changes, airport changes, confirmed cancellations, diversions, and credible connection risks as immediate alert candidates. Treat departure gate changes as alert candidates only inside 24 hours before departure. Treat arrival gate changes as alert candidates only when they affect a connection or pickup. Treat equipment changes as alert candidates only when they affect cabin, seat, or itinerary facts. Keep final landing silent unless the user asked for landing notification or a pickup or connection impact remains. Persist the fingerprint before sending a message. Add one Tracking activity before sending a message. Do not add another activity or message for the same fingerprint. Keep unchanged checks silent. After three consecutive expected source failures, add one degraded-coverage activity. Do not treat a failed read as evidence that nothing changed. Remove the operational job when the Tracking item is completed or retired.

Find flights in the air

Use search to find airborne flights by origin, destination, or area. Default to one page of results and ask before pulling more.

Airport activity

Use airport for an airport's details, airport-delays for current delays, airport-flights for arrivals and departures, and airport-weather for conditions. nearby-airports finds airports around a location.

Airline activity

Use operator for an airline's details and operator-flights for its recent and scheduled flights.

History

For past flights, use the history commands with an explicit date range. They cover flights, tracks, routes, airport activity, and an aircraft's last flight.

Predictions and schedules

Use foresight for FlightAware's predicted status and positions, schedules for scheduled flights between two dates, and disruptions for cancellation and delay counts.

Other commands

The flows above cover the common cases. For anything else, run flightaware --help for the full command list and flightaware <command> --help for a command's options. This includes aircraft owner and type lookups, route and count queries, and advanced search syntax.

Rules

  • Use the airline's ICAO flight number when you can (UAL123, not UA123). If the user's flight number is ambiguous, resolve it with canonical-flight first.
  • For "where is my flight", get the flight first, pick the right date and leg, then look up its position or track. Do not guess an fa_flight_id.
  • A cancellation is a high-impact terminal claim. Never report it as confirmed or stop a flight watch from cancelled: true alone. If the flight carries muse_cancellation_evidence.classification: conflicting_provider_fields, recheck the exact dated leg. If the fields still conflict, seek confirmation from an airline or airport source; otherwise say the cancellation is unconfirmed and keep monitoring. Only treat the cancellation as confirmed when FlightAware's boolean and status text agree, or an independent source confirms it.
  • If FlightAware says a flight or aircraft is blocked or has no data, tell the user plainly and do not try to work around it.
  • Flight reads preserve AeroAPI's raw times and add semantic UTC/user-local fields such as scheduled_gate_departure_at, estimated_takeoff_at, actual_landing_at, and scheduled_gate_arrival_at. Prefer their user_local values in replies. Position samples similarly add position_observed_at.
  • FlightAware does not cover airline policies, terminal maps, baggage, booking, or customer service. For those, use web search and make clear which details came from the web rather than FlightAware.
  • Reply in plain language about the flight, not about how you looked it up. Do not show the user commands, ids, tokens, or raw status codes.

Limits

  • You can't book travel, buy tickets, change a reservation, or contact an airline or airport.
  • Filing a flight intent changes state in FlightAware, so only do it when the user clearly asks and the exact flight is unambiguous; no additional confirmation is required.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Flightaware AI skill do?

Use for questions about a specific flight’s departure or arrival time, including “when’s my flight?” and confirmation of remembered times, plus flight status, delays, and cancellations. Verify the exact dated flight before answering; memory identifies the itinerary but does not verify its current schedule. Use FlightAware to monitor operational changes for an upcoming booked flight when the user directly asks for ongoing monitoring.

Why use Flightaware on TypingMind?

Because you install it once and use it with any model. Flightaware is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Flightaware in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/win4r/MuseAI-Skills/tree/main/opt/hatch/skills/flightaware. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Flightaware?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Flightaware?

As many as you like. As long as a model supports skills, you can use Flightaware with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Flightaware AI skill free?

It is published on GitHub by win4r. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇