consulting-an-oracle

Use when stuck after multiple debug attempts and want to escalate to a stronger one-shot model (GPT-5 Pro, Opus, Gemini Pro) — packages a self-contained "oracle prompt" with Ruby/Rails project briefing, verbatim error, what-was-tried, constraints, and just-enough attached files. Triggers include "ask the oracle", "write a letter to GPT-5", "I'm stuck, draft a prompt for another model", "/tmp/letter.md".

Quality

83%

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

Quality

Content

71%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.

A well-structured, highly actionable skill body with a clean step sequence and concrete code, but it is undercut by a missing bundle file: the core writing step and the closing reference both point to template.md, which is not shipped, breaking progressive disclosure and leaving the final action under-specified.

Suggestions

Ship template.md in the skill bundle (e.g. alongside SKILL.md or under references/) so the Step 6 reference and the closing 'See template.md' pointer resolve to a real file — this is the single highest-impact fix for both progressive_disclosure and actionability.

If template.md is intentionally omitted, inline the oracle-prompt section layout (role / desired output / project briefing / verbatim error / what-was-tried / constraints / attached files / suspect list / safety footer) directly in SKILL.md so the write step is self-contained.

Tighten the design-rationale paragraphs (the model-agnosticism justification and the redaction-timing rationale) to one line each; the 'why' is useful but currently over-explained for a reader who already understands the tradeoffs.

Dimension	Reasoning	Score
Conciseness	The body is lean for a 7-step process and assumes Claude's competence (no basic-concept padding), but a few design-rationale paragraphs — e.g. 'This separation is deliberate: keeping prompt-construction decoupled...keeps the skill model-agnostic...avoids coupling the skill to credentials, CLI availability, or cost decisions that belong to the user' — could be trimmed without losing guidance.	4 / 5
Actionability	Most steps are fully executable — copy-paste bash detection block, concrete redaction regex, hard caps (≤8 files / ≤2000 lines), and a concrete output/invocation template — but the culminating Step 6 ('Use the structure in template.md') defers the core write action to a file that is not present in the bundle, leaving a gap.	4 / 5
Workflow Clarity	A clear 7-step sequence reinforced by a Quick Reference checklist, with an explicit ordering checkpoint ('This step must run before Step 6, not after') for redaction; held back from 5 by the absence of an explicit error-recovery feedback loop and Step 6's reliance on the missing template.	4 / 5
Progressive Disclosure	The reference to template.md is correctly designed as a one-level-deep, clearly signaled split ('Use the structure in template.md', 'See template.md for the exact section layout'), but template.md is absent from the bundle, so the navigation it promises is broken.	3 / 5
	Total	15 / 20 Passed

Description

95%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.

A strong, specific description that clearly states what the skill produces and when to use it, with rich natural trigger phrases and low conflict risk. The only minor gap is that the action surface is essentially one verb ('packages') with enumerated contents rather than a breadth of distinct actions.

Dimension	Reasoning	Score
Specificity	The description lists several concrete components of the packaged artifact ('Ruby/Rails project briefing, verbatim error, what-was-tried, constraints, and just-enough attached files'), but centers on a single action ('packages a self-contained oracle prompt') rather than multiple distinct actions, so it falls just short of comprehensive.	4 / 5
Completeness	It explicitly answers both what ('packages a self-contained oracle prompt with...') and when ('Use when stuck after multiple debug attempts and want to escalate...') with concrete trigger phrases.	5 / 5
Trigger Term Quality	It provides multiple natural trigger phrases users would actually say ('ask the oracle', 'write a letter to GPT-5', 'I'm stuck, draft a prompt for another model'), synonyms across oracle/GPT-5/another model, and a file-path trigger ('/tmp/letter.md').	5 / 5
Distinctiveness Conflict Risk	It occupies a clear niche — escalating stalled Ruby/Rails debugging to a stronger one-shot model — with distinct triggers unlikely to fire for unrelated skills.	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: lucianghinda/superpowers-ruby
Path: skills/consulting-an-oracle/SKILL.md
Commit: 99fac3d

Reviewed: 5 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.