Agent skill

Duffel Flight Booking

by win4r in win4r/MuseAI-Skills

Defines how an agent searches, books and manages flights through the Duffel CLI, and watches a booked fare when you ask for ongoing price monitoring.

No licenceAuto-check passedProductivity & Automation

Install Duffel Flight Booking

skills CLI
$ npx skills add win4r/MuseAI-Skills --skill duffel -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install win4r/MuseAI-Skills duffel --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/win4r/MuseAI-Skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/opt/hatch/skills/duffel .claude/skills/duffel && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
duffel
GitHub stars
332
Used in
1 other repo
Token cost
~7k tokens
SKILL.md length
3,655 words
Files
3
Skills in repo
25
Repo updated
First seen
Licence
None found

At a glance

Defines how an agent searches, books and manages flights through the Duffel CLI, and watches a booked fare when you ask for ongoing price monitoring.

  • Works in 8 steps: Start by searching the complete… → Present the shortlist through → If Travel Planning delegated the search… → …
  • Searching flights for a full itinerary with a specific mix of passengers
  • SKILL.md covers Booking flow, Search and compare, Tracked fare monitoring and Seats, plus 6 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

This skill sets out how an agent uses the Duffel flight CLI to search for flights, show seat options, validate a booking and book it. It can also watch the fare on a flight you already booked when you directly ask for ongoing price monitoring. It is written for the Muse AI runtime and defers to a separate booking skill and its reference files for the rules on discovery, privacy, presentation and commitment; this file covers only the Duffel command contract.

Searches cover the full itinerary and exact passenger mix together and are saved straight to separate temporary JSON files, never piped through filters, so errors and offer data stay visible. The saved file is then inspected read-only to rank offers and reused for presentation. The agent does not invent identifiers, call undocumented endpoints, show Duffel commands or internal offer and order IDs, or ask for payment-card data or account credentials. After you pick an itinerary it can show seats, collect passenger details and any loyalty account, then book natively or through the airline website.

When a travel-planning step delegates a search in a planning-only posture, the agent returns the chosen itinerary and stops until the posture changes to ready to book. The folder includes a manifest and a YAML file of tracked-flight test scenarios, and the tool list covers file access, tracking, user-goal and scheduled-job operations.

When your agent uses it

  • Searching flights for a full itinerary with a specific mix of passengers
  • Booking a selected flight after seats and passenger details are collected
  • Monitoring the fare of a flight you already booked when you ask for price tracking

Example prompts

  • “Find round-trip flights from Chicago to Lisbon next month for two adults and one child.”
  • “Show seat options for the flight I just picked and add my frequent flyer number.”
  • “Keep an eye on the fare for my booked Lisbon flight and tell me if it drops.”

Requirements

  • The Duffel CLI as exposed by the host runtime
  • The companion booking skill and its flight reference files
  • Pre-approved tools (allowed-tools): exec, widget.create, read, write, edit, tracking.list, tracking.get, tracking.search, tracking.create, tracking.update, tracking.create_entry, tracking.set_status, user_goal.list, user_goal.get, user_goal.search, user_goal.create_entry, cron.list, cron.view, cron.status, cron.add, cron.update, cron.remove

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

  1. Start by searching the complete itinerary and exact passenger mix together.
  2. Present the shortlist through
  3. If Travel Planning delegated the search in planning only posture, return
  4. Optionally show seat-options and collect bookable seats.
  5. Collect required passenger details and any loyalty account the user wants
  6. Let /opt/hatch/skills/booking/references/flights.md choose between the
  7. Otherwise continue through the supported native path without asking the
  8. On success, report every carrier confirmation reference/PNR and ticket

What it can do on your machine

Read from SKILL.md and the folder at commit 38bbb45. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • exec
    • widget.create
    • read
    • write
    • edit
    • tracking.list
    • tracking.get
    • tracking.search
    • tracking.create
    • tracking.update

    …and 12 more on the same allowed-tools line.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are bash and json).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Duffel Flight Booking loads about 7k tokens when it runs. Until then it costs about 44 tokens; SKILL.md has 3,655 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~44
When it runs · the whole SKILL.md, loaded when a task matches
~7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

Without a licence we can't republish the file, so here is its outline and opening line. It has 3,655 words (~6,976 tokens).

“For a user-facing flight search or booking, first read /opt/hatch/skills/booking/SKILL.md, /opt/hatch/skills/booking/references/flights.md, and /opt/hatch/skills/booking/references/presentation.md. Those files define the discovery, privacy, presentation, and commitment rules. Use this file for the Duffel CLI contract. Do not emit HTML, call create_options, or duplicate flight…”

— opening of SKILL.md by win4r
name
duffel
allowed-tools
exec, widget.create, read, write, edit, tracking.list, tracking.get, tracking.search, tracking.create, tracking.update, tracking.create_entry, tracking.set_status, user_goal.list, user_goal.get, user_goal.search, user_goal.create_entry, cron.list, cron.view, cron.status, cron.add, cron.update, cron.remove
metadata.includeInPrompt
false

Read the full SKILL.md on GitHub

Files

SKILL.md and 2 other files in opt/hatch/skills/duffel of win4r/MuseAI-Skills.

  • SKILL.md
  • eval/tracked-flight-scenarios.yaml
  • manifest.yaml

Open the folder on GitHubat commit 38bbb45

Used in 1 other repository

We found 1 copy of this SKILL.md (exact, near-identical or edited) in other folders, from 1 other GitHub owner. This page covers the copy in win4r/MuseAI-Skills, which our catalogue first saw on October 7, 2026.

Compare with similar skills

Duffel Flight Booking next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Duffel Flight Booking compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Duffel Flight Booking this skillwin4r/MuseAI-Skills3321 repos~7kAutomated safety check: PassNone
ComposioComposioHQ/composio30k1 repos~1.7kAutomated safety check: PassMIT
Effect V4ComposioHQ/composio30k—~1kAutomated safety check: PassMIT
Cap Screen Recording CLICapSoftware/Cap23k—~1.7kAutomated safety check: PassCustom licence
Feishu Bitable Managementop7418/CodePilot6.5k1 repos~1.7kAutomated safety check: PassCustom licence
ScheduleTinyAGI/tinyagi3.6k—~1.4kAutomated safety check: PassMIT

Similar skills

  • Composio

    ComposioHQ/composio

    Route and complete Composio work across Composio For You and Composio Platform.

    30k GitHub starsUsed in 1 repo~1.7k tokens
    Productivity & AutomationAuto-check passed
  • Effect V4

    ComposioHQ/composio

    Write, review, or upgrade Effect v4 code in the Composio CLI, cli-keyring, and json-schema-to-effect-schema packages, all pinned exactly to effect@4.0.0-rc.117 — Context.Service and explicit layers…

    30k GitHub stars~1k tokensUpdated yesterday
    Productivity & AutomationAuto-check passed
  • Operates Cap, the screen recording and sharing tool, through its local MCP tools or cap command instead of browser automation: library, transcripts, comments and sharing.

    23k GitHub stars~1.7k tokensUpdated yesterday
    Productivity & AutomationAuto-check passed
  • Creates, queries and edits Feishu Bitable tables through tool calls, covering records, fields, views and batch operations with the correct value format for each field type.

    6.5k GitHub starsUsed in 1 repo~1.7k tokens
    Productivity & AutomationAuto-check passed
  • Schedule

    TinyAGI/tinyagi

    Create, list, and delete scheduled tasks (recurring or one-time) that send messages to agents.

    3.6k GitHub stars~1.4k tokensUpdated 6 mo ago
    Productivity & AutomationAuto-check passed
  • OpenObserve PR Brief

    openobserve/openobserve

    Produces a read-only morning brief of your open pull requests across the openobserve GitHub org, with a next step for each and a reminder for idle ones.

    22k GitHub stars~1.8k tokensUpdated yesterday
    Productivity & AutomationAuto-check passed

More from win4r/MuseAI-Skills

All 25 skills in this repo
  • Withings Health Data Reader

    win4r/MuseAI-Skills

    Reads Withings body measurements, daily activity, sleep sessions and workouts through a bundled withings command line tool and returns typed JSON records.

    332 GitHub starsUsed in 2 repos~1.3k tokens
    Auto-check passed
  • Travel and Event Booking

    win4r/MuseAI-Skills

    Handles direct flight, hotel, restaurant and event-ticket bookings and live availability checks, loading the right category reference before searching or booking.

    332 GitHub starsUsed in 1 repo~4k tokens
    Auto-check passed
  • HealthEx Health Records

    win4r/MuseAI-Skills

    Connects to HealthEx through OAuth and its MCP server so you can ask about medications, lab results, conditions and other health records.

    332 GitHub starsUsed in 1 repo~3.7k tokens
    Auto-check passed
  • Product Shopping Search

    win4r/MuseAI-Skills

    Guides an agent through product searches across a catalog, a browser agent and Facebook Marketplace, favoring low prices and citing each product it names.

    332 GitHub starsUsed in 1 repo~9.1k tokens
    Auto-check passed
  • Muse Feedback Filer

    win4r/MuseAI-Skills

    Tells the agent when to offer, and how to file, a feature request or bug report with the Muse team through a feature-request command line tool, only with your approval.

    332 GitHub starsUsed in 1 repo~2.7k tokens
    Auto-check: warnings
  • Forget Personal Information

    win4r/MuseAI-Skills

    Removes a personal fact or past event from Muse's memory and from reminders, summaries and scheduled work that could bring it back, after you confirm a plan.

    332 GitHub starsUsed in 1 repo~1.6k tokens
    Auto-check: warnings

Questions about Duffel Flight Booking

What does Duffel Flight Booking do?

Defines how an agent searches, books and manages flights through the Duffel CLI, and watches a booked fare when you ask for ongoing price monitoring. This skill sets out how an agent uses the Duffel flight CLI to search for flights, show seat options, validate a booking and book it. It can also watch the fare on a flight you already booked when you directly ask for ongoing price monitoring.

When should I use Duffel Flight Booking?

Duffel Flight Booking fits situations like: searching flights for a full itinerary with a specific mix of passengers; booking a selected flight after seats and passenger details are collected; monitoring the fare of a flight you already booked when you ask for price tracking.

How do I install Duffel Flight Booking in Claude Code?

Run `npx skills add win4r/MuseAI-Skills --skill duffel -a claude-code`. Or copy the skill folder (opt/hatch/skills/duffel in win4r/MuseAI-Skills) into .claude/skills/duffel in your project. Claude Code loads it when a task matches its description.

How do I install Duffel Flight Booking in Codex?

Run `npx skills add win4r/MuseAI-Skills --skill duffel -a codex`. Or copy the skill folder (opt/hatch/skills/duffel in win4r/MuseAI-Skills) into .agents/skills/duffel in your project. Codex loads it when a task matches its description.

Can I use Duffel Flight Booking in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add win4r/MuseAI-Skills --skill duffel -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/duffel, .gemini/skills/duffel, .github/skills/duffel and .opencode/skills/duffel in your project.

What does Duffel Flight Booking need to run?

SKILL.md names no scripts, command-line tools or credentials: Duffel Flight Booking is instructions for the agent only. Our summary lists: The Duffel CLI as exposed by the host runtime; The companion booking skill and its flight reference files. Its frontmatter pre-approves these tools: exec, widget.create, read, write, edit, tracking.list, tracking.get, tracking.search, tracking.create, tracking.update, tracking.create_entry, tracking.set_status, user_goal.list, user_goal.get, user_goal.search, user_goal.create_entry, cron.list, cron.view, cron.status, cron.add, cron.update, cron.remove.

Does Duffel Flight Booking access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Duffel Flight Booking safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Duffel Flight Booking use?

No licence was found for Duffel Flight Booking or its repository. Without one, default copyright applies: ask the author before reusing or redistributing it.

How many tokens does Duffel Flight Booking use?

About 7k tokens (SKILL.md is roughly 28k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Duffel Flight Booking?

Skills that share tags, products or a category with Duffel Flight Booking: Composio (ComposioHQ/composio, 30k stars), Effect V4 (ComposioHQ/composio, 30k stars), Cap Screen Recording CLI (CapSoftware/Cap, 23k stars) and Feishu Bitable Management (op7418/CodePilot, 6.5k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Duffel Flight Booking?

win4r (a GitHub user) maintains it in win4r/MuseAI-Skills, which has 332 GitHub stars. The repository holds 25 skills in this directory. The repository was last updated on September 28, 2026.

Source: win4r/MuseAI-Skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.