ralph

Self-referential loop until task completion with configurable verification reviewer

Quality

56%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/ralph/SKILL.md

Quality

Content

70%Weight 40%Scale 1-5

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

The content is a well-sequenced, highly actionable orchestration workflow with strong validation checkpoints and feedback loops. Its main weaknesses are redundant warnings that hurt conciseness and references to docs/ files that are not present in the bundle, leaving progressive disclosure only partially realized.

Suggestions

Dedupe the 'do not stop after Step 7 approval / polite-stop anti-pattern' warning and the ai-slop-cleaner skill-vs-agent note so each appears once, tightening conciseness toward 4-5.

Move the large PRD_Mode / Execution_Policy / company-context-interface detail into referenced files under references/ and link to them from the body, and ensure the referenced docs/ paths actually exist in the bundle so progressive disclosure resolves.

Consolidate the agent-tier table and reviewer-selection rules into a single reference section to reduce inline bulk.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient procedural guidance, but repeats the same warnings in multiple places (the 'do not stop after Step 7 approval' / polite-stop anti-pattern in Step 7 and again in Escalation_And_Stop_Conditions; the ai-slop-cleaner skill-vs-agent warning in Step 7.5 and Tool_Usage), so it could be tightened rather than earning 4.	3 / 5
Actionability	Guidance is largely executable with concrete Task() calls carrying explicit model tiers, specific commands (omc ask codex --agent-prompt critic, Skill("ai-slop-cleaner"), /oh-my-claudecode:cancel), and specific file paths, with only minor illustrative/conditional gaps keeping it below 5.	4 / 5
Workflow Clarity	A clearly sequenced 1-9 step process with explicit validation checkpoints (Step 4 acceptance-criteria verification, Step 7 reviewer verification, Step 7.6 regression re-verification), rejection feedback loops (Step 9), and a Final_Checklist, matching the top anchor including checklists and error-recovery loops.	5 / 5
Progressive Disclosure	There is section structure and references to docs/ paths are signaled (docs/shared/agent-tiers.md, docs/company-context-interface.md, docs/REFERENCE.md), but no bundle files exist so those references cannot resolve, and substantial detail (PRD_Mode, Execution_Policy, company-context interface) is inlined that could live in separate files, fitting the 'some structure, references not fully resolvable, content that should be separate is inline' anchor.	3 / 5
	Total	15 / 20 Passed

Description

42%Weight 40%Scale 1-5

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 conveys what Ralph does with a couple of concrete mechanisms but leans on jargon and omits any explicit 'use when' trigger guidance. It is moderately specific and moderately distinct, with natural trigger terms being the weakest aspect.

Suggestions

Add an explicit 'Use when...' clause with natural trigger phrases users would say (e.g. 'Use when the user says ralph, don't stop, keep going until done, or must complete with verification').

Replace jargon like 'self-referential loop' and 'configurable verification reviewer' with plainer user-facing terms and synonyms to improve trigger-term quality.

Enumerate a few more concrete actions or outcomes to lift specificity toward listing multiple specific behaviors.

Dimension	Reasoning	Score
Specificity	Names the domain and a couple of concrete mechanisms ('Self-referential loop until task completion', 'configurable verification reviewer') but does not enumerate multiple specific actions, so it sits at the 'domain + 1-2 actions, not comprehensive' anchor rather than 4.	3 / 5
Completeness	The 'what' is clear but there is no 'Use when...' clause or equivalent explicit trigger guidance, so per the rubric guideline completeness is capped at 3 and does not reach 4.	3 / 5
Trigger Term Quality	The terms are technical ('self-referential loop', 'configurable verification reviewer') and miss the natural phrases a user would actually say ('don't stop', 'keep going until done', 'finish this'), matching the 'one or two generic keywords; missing natural phrases' anchor rather than 3.	2 / 5
Distinctiveness Conflict Risk	The self-referential-loop-plus-reviewer framing is somewhat specific but 'loop until task completion' is broad enough to overlap with other persistence/loop skills, fitting the 'somewhat specific but could still overlap' anchor instead of 4.	3 / 5
	Total	11 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: Yeachan-Heo/oh-my-claudecode
Path: skills/ralph/SKILL.md
Commit: 5aa678c

Reviewed: 8 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.