ralph
Self-referential loop until task completion with configurable verification reviewer
50
38%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/ralph/SKILL.mdQuality
Discovery
0%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 is too abstract and technical to be useful for skill selection. It fails to explain what specific tasks the skill handles, uses jargon instead of natural user language, and provides no guidance on when Claude should invoke it. The description would benefit from concrete examples of use cases and explicit trigger conditions.
Suggestions
Replace abstract language with concrete actions - specify what types of tasks this skill iterates on (e.g., 'Iteratively refines code, documents, or analysis until quality criteria are met').
Add a 'Use when...' clause with natural trigger terms users would say, such as 'keep trying', 'iterate until done', 'verify and fix', 'don't stop until complete'.
Clarify what 'configurable verification reviewer' means in practical terms - what gets verified and how does the user configure it?
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description uses vague, abstract language like 'self-referential loop' and 'task completion' without specifying what concrete actions the skill performs or what types of tasks it handles. | 1 / 3 |
Completeness | The description vaguely hints at 'what' (some kind of looping/verification process) but provides no 'when' clause or explicit triggers for when Claude should select this skill. | 1 / 3 |
Trigger Term Quality | Contains technical jargon ('self-referential loop', 'configurable verification reviewer') that users would not naturally say when seeking help. No natural keywords like 'iterate', 'repeat', 'verify', or 'check work' are included. | 1 / 3 |
Distinctiveness Conflict Risk | The phrase 'task completion' is extremely generic and could apply to virtually any skill. There are no distinct triggers that would help Claude differentiate this from other skills. | 1 / 3 |
Total | 4 / 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, highly actionable skill for a complex persistence loop workflow. Its strengths are excellent workflow clarity with explicit validation checkpoints and concrete, executable guidance. Weaknesses include moderate verbosity (explanatory sections that could be trimmed) and a monolithic structure that could benefit from splitting into multiple files.
Suggestions
Trim or remove the 'Why_This_Exists' section - Claude doesn't need motivation, just instructions
Move the Examples section to a separate EXAMPLES.md file and reference it, reducing the main skill length
Condense the Good/Bad examples - they repeat concepts already covered in Steps
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is moderately verbose with some redundancy (e.g., Good/Bad examples repeat concepts already explained in Steps). The Purpose, Use_When, and Do_Not_Use_When sections are useful but could be tighter. Some explanations like 'Why_This_Exists' add context Claude likely doesn't need. | 2 / 3 |
Actionability | Provides highly concrete guidance with specific Task() syntax, exact file paths (prd.json, progress.txt), explicit commands (/oh-my-claudecode:cancel), and detailed examples showing correct vs incorrect patterns. The Steps section is executable and copy-paste ready. | 3 / 3 |
Workflow Clarity | Excellent multi-step workflow with clear sequencing (Steps 1-9), explicit validation checkpoints (Step 4 verification, Step 7 reviewer verification, Step 7.6 regression re-verification), and feedback loops (rejection → fix → re-verify). The Final_Checklist provides comprehensive validation. | 3 / 3 |
Progressive Disclosure | Content is mostly inline with one reference to external docs (docs/shared/agent-tiers.md). The skill is long (~200 lines) and could benefit from splitting Examples, Advanced, and Tool_Usage into separate files. However, sections are well-organized with clear headers. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- Yeachan-Heo/oh-my-claudecode
- Commit
48ffaac
- Reviewed
Table of Contents
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.