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
dld-common
skills/dld-common/SKILL.md
Discovery
32%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description serves a unique purpose as an internal/shared utility skill, which explains its lack of user-facing trigger terms. However, it fails to specify what utilities it provides or which DLD skills depend on it, making it difficult for maintainers to understand its scope. The explicit non-invocation statement is appropriate for its purpose but the description would benefit from listing its actual capabilities.
Suggestions
Add specific capabilities listing what utilities are provided (e.g., 'Provides file parsing helpers, date formatting, and validation functions').
Specify which DLD skills depend on this utility or what category of functionality it supports to help with maintenance and understanding.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description uses vague language like 'utility scripts' without specifying what actions these scripts perform. No concrete capabilities are listed. | 1 / 3 |
Completeness | The 'what' is weakly addressed (shared utilities), and the 'when' is actually a 'when NOT to use' statement. It explains it's for internal use but lacks explicit trigger guidance for when other skills should invoke it. | 2 / 3 |
Trigger Term Quality | Contains only technical jargon ('DLD skills', 'utility scripts') with no natural keywords a user would say. The description explicitly states it's not for direct invocation, so trigger terms are intentionally absent. | 1 / 3 |
Distinctiveness Conflict Risk | The explicit statement 'Not intended for direct invocation' clearly distinguishes this from user-facing skills, making it unlikely to conflict with other skills in selection scenarios. | 3 / 3 |
Total | 7 / 12 Passed |
Implementation
37%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is extremely concise but lacks actionability. While it correctly identifies itself as an internal utility not meant for direct invocation, it provides no guidance on how the scripts should be called, what parameters they accept, or what outputs to expect. For a shared utility skill, it should at minimum document the interface for each script.
Suggestions
Add usage examples or parameter signatures for each script (e.g., `scripts/update-status.sh <decision-id> <new-status>`)
Document expected outputs or return values for scripts like `next-id.sh`
Add brief notes on which DLD skills depend on these utilities, or link to them
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely lean and efficient. No unnecessary explanation of what scripts are or how they work. Every line serves a purpose - just lists what exists and what it does. | 3 / 3 |
Actionability | Provides no concrete guidance on how to use these scripts. No examples of invocation, no parameters shown, no expected outputs. Just a list of script names with brief descriptions. | 1 / 3 |
Workflow Clarity | No workflow information provided. While the skill states it's 'not intended for direct invocation,' there's no guidance on how other skills should use these utilities or in what sequence. | 1 / 3 |
Progressive Disclosure | Simple structure appropriate for a utility skill, but lacks any references to where these scripts are actually used or how to find documentation on their parameters/usage within the scripts themselves. | 2 / 3 |
Total | 7 / 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.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.