ledger-dmk-implementation

Execute Ledger Device Management Kit (DMK) signing operations safely: initialize the SDK, establish a device session, verify device state, open the correct chain app, perform signing or device management operations, and return the result. Use when a developer needs to implement a signing operation (transaction, message, typed data), derive or retrieve an address from a Ledger device, send any command to a Ledger device, or perform device management operations such as genuine check, app install, or app uninstall.

1.38x

Quality

—

Does it follow best practices?

Impact

98%

1.38x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Quality

Content

77%

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

An excellently sequenced, validation-heavy signing workflow with strong gate discipline and error handling. Its main weaknesses are that complete executable code lives only in referenced files and that those referenced bundle files are absent from the skill bundle, undermining progressive disclosure.

Suggestions

Ship the referenced bundle files (dmk-sdk-reference.md, dmk-code-patterns.md, dmk-platform-patterns.md) under references/ so the signaled progressive-disclosure structure is real rather than dangling.

Inline at least one minimal end-to-end executable code block (e.g. a full ETH sign example with imports) in SKILL.md so the body is copy-paste ready, not just API-call fragments.

Confirm the cross-referenced sibling skills (dmk-intent-vocabulary, dmk-business-logic) resolve to actual skills, or note them as optional external dependencies.

Dimension	Reasoning	Score
Conciseness	Dense but earned: terse gate notation (→ PROCEED/WAIT/ABORT/ESCALATE) and state tables convey a complex flow without explaining concepts Claude already knows, and every explanatory note (e.g. "the session is a transport connection, not an authorization") carries operational value rather than padding.	3 / 3
Actionability	Concrete API calls and commands appear inline ("dmk.startDiscovering({ transport: webHidIdentifier })", "signPsbt()", "dmk.executeDeviceAction()", "sendCommand(CloseAppCommand)") but complete, copy-paste-ready code is deferred to dmk-code-patterns.md rather than present in the body, leaving the inline guidance real but incomplete for direct execution.	2 / 3
Workflow Clarity	Steps 1–5 are explicitly sequenced as gates, each with a validation checkpoint (session-state checks, app-state table), plus error-classification and HITL escalation checklists that provide clear feedback loops — matching the explicit-validation anchor.	3 / 3
Progressive Disclosure	Reference files are well-signaled with load triggers and kept one level deep, but the referenced bundle files (dmk-sdk-reference.md, dmk-code-patterns.md, dmk-platform-patterns.md) are not present in the bundle — references/ and other bundle directories do not exist — so the disclosure structure cannot actually deliver the split it promises.	2 / 3
	Total	10 / 12 Passed

Description

100%

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

A tightly written, third-person description that states concrete capabilities and pairs them with an explicit, multi-trigger 'Use when' clause. It is specific, complete, and clearly distinguishable from other skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "initialize the SDK, establish a device session, verify device state, open the correct chain app, perform signing or device management operations, and return the result" — matching the multi-specific-actions anchor. Voice is third-person imperative ("Execute..."), so no penalty applies.	3 / 3
Completeness	Explicitly answers both what (the full safe signing flow) and when via an explicit "Use when a developer needs to..." trigger clause, matching the top anchor.	3 / 3
Trigger Term Quality	Covers natural developer phrasing — "signing operation", "transaction, message, typed data", "derive or retrieve an address", "send any command to a Ledger device", "genuine check, app install, app uninstall" — broad coverage of terms a user would actually say.	3 / 3
Distinctiveness Conflict Risk	Niche is sharply scoped to Ledger DMK signing/device-management operations with distinct triggers, making overlap with unrelated skills unlikely.	3 / 3
	Total	12 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: LedgerHQ/agent-skills
Commit: c9739bb

Reviewed: 6 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.