Name: duffel
Rating: 60.8 (1 reviews)
Author: membranedev

Blog Docs Log in Get started

duffel

Duffel integration. Manage data, records, and automate workflows. Use when the user wants to interact with Duffel data.

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/duffel/SKILL.md

Quality

Content

81%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is a tight, executable guide with a strong validation-rich connection workflow and clean sectioning. Its main weaknesses are minor placeholder gaps in command examples and the absence of one concrete end-to-end Duffel action demonstration.

Suggestions

Trim or remove the opening background paragraph about what Duffel is and who uses it, since Claude already knows this.

Add one fully worked example that substitutes real values, e.g. a sample intent and a concrete `membrane action run` call with a populated actionId and input, so the run flow is copy-paste-complete.

Consider moving the lengthy connection-state enumeration into a short reference table or a referenced file to keep the main flow scannable.

Dimension	Reasoning	Score
Conciseness	The body is largely efficient and command-driven, with only minor over-explanation such as the opening paragraph describing what Duffel is and who uses it.	4 / 5
Actionability	It provides copy-paste-ready, executable commands for install, auth, connection, searching, running actions, and proxy requests, with minor gaps from unfilled placeholders (CONNECTION_ID, QUERY intent) and no concrete end-to-end Duffel action example.	4 / 5
Workflow Clarity	The connection-readiness workflow is explicitly sequenced with a poll-until-READY feedback loop and enumerated recovery paths for BUILDING, CLIENT_ACTION_REQUIRED, and CONFIGURATION_ERROR states.	5 / 5
Progressive Disclosure	The document is well-organized into clear sections with no need for external bundle files, though its length and the inlined connection-state enumeration leave minor organization gaps versus a fully split structure.	4 / 5
	Total	17 / 20 Passed

Description

62%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description covers both what and when and is anchored to the distinct Duffel brand, but its capabilities are generic and its trigger terms miss the natural flight-booking vocabulary users would actually say. Tightening the actions to Duffel-specific operations and adding concrete trigger phrases would lift the weaker dimensions.

Suggestions

Replace generic actions ('Manage data, records, automate workflows') with Duffel-specific ones, e.g. 'Search flights, create orders, manage bookings and refunds via the Duffel API'.

Add concrete trigger terms users naturally say: 'Use when the user wants to search or book flights, manage airline orders, issue tickets, or handle refunds through Duffel.'

Avoid the shared Membrane boilerplate phrasing so the description reads as Duffel-specific rather than a template applied to many integrations.

Dimension	Reasoning	Score
Specificity	It names the domain and lists several actions ('Manage data, records, and automate workflows'), but those actions are generic rather than concrete Duffel-specific operations like flight search, ordering, or issuing tickets.	3 / 5
Completeness	Both 'what' ('Manage data, records, and automate workflows') and 'when' ('Use when the user wants to interact with Duffel data') are present, though the 'when' is generic rather than a concrete trigger phrase.	4 / 5
Trigger Term Quality	The relevant keyword is the brand 'Duffel' ('interact with Duffel data'), but it omits the natural domain terms users would say such as flights, flight search, booking, orders, or airlines.	3 / 5
Distinctiveness Conflict Risk	The 'Duffel' brand anchor gives it a distinct niche with minimal conflict risk, though the boilerplate 'manage data, records, automate workflows' phrasing overlaps with sibling Membrane integration skills.	4 / 5
	Total	14 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: membranedev/application-skills
Path: skills/duffel/SKILL.md
Commit: f484c82

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.