Content
22%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill claims to validate Conductor project artifacts but fails to provide a coherent validation workflow with pass/fail criteria, error handling, or recovery steps. The content is a mix of concrete but incomplete shell commands and generic boilerplate instructions that don't teach Claude anything useful. The pattern matching section is the strongest part but is disconnected from any validation logic.
Suggestions
Define a clear sequential validation workflow: check directory exists → verify required files → validate file contents → report results with pass/fail status for each check.
Replace generic Instructions bullets ('Apply relevant best practices') with specific validation checks, e.g., 'Verify tracks.md contains at least one track entry matching the pattern <type>_<name>_<YYYYMMDD>'.
Add explicit error handling: what should Claude do when a required file is missing or a pattern doesn't match? Include concrete remediation steps or error messages.
Remove the boilerplate 'Use this skill when / Do not use this skill when' sections or make them specific to actual validation scenarios rather than repeating 'check if conductor directory exists'.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is padded with generic boilerplate ('Clarify goals, constraints, and required inputs', 'Apply relevant best practices') that adds no value. The 'Use this skill when' and 'Do not use this skill when' sections repeat the same phrase ('check if conductor directory exists') multiple times without adding clarity. The limitations section is entirely generic filler. | 1 / 3 |
Actionability | The initial shell commands (ls -la conductor/, checking required files) are concrete and executable, and the pattern matching section provides specific marker formats. However, the Instructions section is entirely vague ('Apply relevant best practices and validate outcomes') with no concrete validation logic, error handling, or what to do when files are missing. | 2 / 3 |
Workflow Clarity | There is no clear sequenced workflow for validation. The skill lists some ls commands at the top but doesn't define what to do with results, how to handle missing files, or what constitutes a passing vs failing validation. For a validation skill, the absence of any pass/fail criteria or error recovery steps is a critical gap. | 1 / 3 |
Progressive Disclosure | There is one reference to 'resources/implementation-playbook.md' for detailed examples, which is appropriate. However, no bundle files exist to support this reference, and the main content mixes overview-level commands with pattern-matching details that could be better organized. | 2 / 3 |
Total | 6 / 12 Passed |