prometheus-strict

[OMX] Clean-room interview-driven planner: Metis clarifies, Momus challenges, Oracle synthesizes, then hands off to $ultragoal/$team.

Quality

48%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./plugins/oh-my-codex/skills/prometheus-strict/SKILL.md

Quality

Content

62%

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

The body excels at workflow clarity with explicit sequencing, validation checkpoints, bounded feedback loops, and checklists. It is weakened by verbosity and repetition, partially-abstract role guidance, and a monolithic structure with no reference files despite being long enough to warrant them.

Suggestions

Extract the Metis, Momus, and Oracle role definitions into separate reference files (e.g., references/metis.md) referenced one level deep, reducing the inline monolith.

Deduplicate the research fan-out and gap-fill guidance so each appears once and cross-references rather than repeating across Execution_Policy, Steps, and Tool_Usage.

Tighten the role descriptions into concrete, callable instructions so the core workflow is as executable as the handoff commands and output contract already are.

Dimension	Reasoning	Score
Conciseness	The body is information-dense and avoids explaining basic known concepts, but at ~218 lines it repeats the same mechanics (research fan-out, gap-fill passes) across Execution_Policy, Steps §2, and Tool_Usage, and could be tightened considerably.	2 / 3
Actionability	Concrete elements are present (omx question, OMX_QUESTION_RETURN_PANE=$TMUX_PANE, $ultragoal/$team handoff commands, the output-contract template), but the core Metis/Momus/Oracle workflow is described as role abstractions rather than fully executable, copy-paste-ready instructions.	2 / 3
Workflow Clarity	The §1–§6 sequence is explicit with validation checkpoints (6-item checklist clearance, two gap-fill passes, Pass 1↔Pass 2 self-verification), bounded retry/feedback loops (3 Momus→Oracle cycles, 5-round cap), a Final_Checklist, and clear escalation rules.	3 / 3
Progressive Disclosure	No bundle files exist (references/scripts/assets are absent) and the 218-line body is a monolith; the Metis/Momus/Oracle role definitions and output contract are inline content that would benefit from being split into one-level-deep reference files.	2 / 3
	Total	9 / 12 Passed

Description

35%

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 a distinctive multi-role planning concept but leans on internal OMX jargon and omits any explicit 'Use when' trigger guidance. It answers 'what' adequately while leaving 'when' implicit, which limits trigger-term quality and completeness.

Suggestions

Add an explicit 'Use when...' clause with natural-language triggers (e.g., 'Use when a task is high-stakes, requirements are ambiguous, and a strict interview before execution is wanted').

Replace or gloss jargon role names with user-facing language so natural keywords surface (e.g., 'clarifies requirements, challenges assumptions, synthesizes a handoff-ready plan').

Make the actions concrete by naming their objects (e.g., 'clarifies scope and acceptance criteria, challenges assumptions and validation gaps').

Dimension	Reasoning	Score
Specificity	The phrase 'Metis clarifies, Momus challenges, Oracle synthesizes, then hands off to $ultragoal/$team' names a domain (interview-driven planner) and several actions, but the verbs (clarifies/challenges/synthesizes) act on unspecified objects, so they stay abstract rather than the concrete, multi-action list of the top anchor.	2 / 3
Completeness	It states what the skill does (clarify/challenge/synthesize/handoff) but has no 'Use when...' clause or explicit trigger guidance, so the 'when' is missing rather than clearly answered — capping completeness at 2 per the guidelines.	2 / 3
Trigger Term Quality	Terms like '[OMX]', 'Clean-room interview-driven planner', 'Metis', 'Momus', 'Oracle', and '$ultragoal/$team' are internal jargon a user would not naturally say; there are no common natural-language trigger keywords.	1 / 3
Distinctiveness Conflict Risk	The OMX planning niche is fairly specific, but the body itself acknowledges overlap with existing $plan, $ralplan, and $deep-interview skills, so it could still be selected for the wrong planning skill.	2 / 3
	Total	7 / 12 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-codex
Path: plugins/oh-my-codex/skills/prometheus-strict/SKILL.md
Commit: 435d4a9

Reviewed: 2 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.