CtrlK
BlogDocsLog inGet started
Tessl Logo

dld-kit/dld

Decision-Linked Development (DLD) — a workflow for recording, linking, and maintaining development decisions alongside code. Skills for planning, recording, implementing, auditing, and documenting decisions via @decision annotations.

68

Quality

68%

Does it follow best practices?

Impact

Pending

No eval scenarios have been run

Overview
Skills
Evals
Files

dld-audit-auto

skills/dld-audit-auto/SKILL.md

Discovery

50%

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 communicates the autonomous, CI-focused nature of the skill and mentions key actions, but remains too vague about what specifically is being audited and what types of drift/issues it handles. It lacks an explicit 'Use when...' clause and could benefit from more specific trigger terms that users would naturally use.

Suggestions

Add an explicit 'Use when...' clause specifying triggers like 'scheduled runs', 'CI pipeline', 'automated maintenance', or 'unattended execution'.

Specify what is being audited (e.g., configuration files, infrastructure, dependencies, code standards) to improve specificity and reduce conflict risk.

Include more natural trigger terms users might say, such as 'cron', 'automated checks', 'headless mode', or 'batch processing'.

DimensionReasoningScore

Specificity

Names domain (audit) and some actions (detects drift, fixes issues, opens a PR), but lacks comprehensive detail about what types of drift, what issues, or what systems are being audited.

2 / 3

Completeness

Answers 'what' (detects drift, fixes issues, opens PR) and implies 'when' (scheduled/CI execution), but lacks an explicit 'Use when...' clause with clear trigger guidance.

2 / 3

Trigger Term Quality

Includes some relevant terms like 'audit', 'drift', 'PR', 'CI', 'scheduled', but missing common variations users might say like 'automated checks', 'cron job', 'pipeline', 'configuration drift', or 'unattended mode'.

2 / 3

Distinctiveness Conflict Risk

The combination of 'autonomous', 'audit', 'drift', and 'CI execution' provides some distinctiveness, but 'audit' and 'fixes issues' are generic enough to potentially overlap with other code quality or maintenance skills.

2 / 3

Total

8

/

12

Passed

Implementation

77%

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

This is a well-structured autonomous workflow skill with excellent actionability and workflow clarity. The step-by-step process is clear with appropriate validation checkpoints and decision trees for handling edge cases. Minor improvements could be made in conciseness (some explanatory text could be trimmed) and progressive disclosure (detailed fix logic could be externalized for very long sections).

DimensionReasoningScore

Conciseness

The skill is mostly efficient but includes some redundant explanations (e.g., explaining what each issue category means when Claude would understand from context). The PR template and fix descriptions could be more condensed.

2 / 3

Actionability

Provides concrete, executable bash commands throughout, specific git operations, clear script paths, and detailed PR template structure. The fix categories have explicit instructions for what actions to take.

3 / 3

Workflow Clarity

Excellent step-by-step sequencing with clear prerequisites, numbered steps, and explicit validation (clean working tree check, commit reachability verification). Includes decision logic for edge cases and clear separation between auto-fixable vs. best-effort fixes.

3 / 3

Progressive Disclosure

Content is well-organized with clear sections, but the skill is somewhat monolithic. The detailed fix logic for each category could potentially be split into referenced files. Script paths are listed but the actual script contents are external, which is appropriate.

2 / 3

Total

10

/

12

Passed

Validation

100%

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

Validation11 / 11 Passed

Validation for skill structure

No warnings or errors.