agentmods.dev

Prometheus-MCP: Skill for OpenCode

.opencode/skills/prometheus-critique/SKILL.md

prometheus-critique is a skill for OpenCode from gguatit/Prometheus-MCP. It costs 169 tokens per session (1,086 once invoked), scanned A, original, MIT.

A review guide for checking and improving creative technical work such as web interfaces, 3D scenes, shaders, animations, and game-feel code. It uses the Prometheus MCP server to evaluate an artifact against defined quality rules.

In plain words
What is it for?
Use it after creating or changing HTML/CSS/JSX interfaces, Three.js or WebGL scenes, shaders, animations, or game code that needs critique or quality checking.
Why use it?
It provides structured feedback for visual and interactive work instead of relying only on the creator's own inspection.

Skill for OpenCode

Written for OpenCode: installed under .opencode/.

This is gguatit/Prometheus-MCP's own configuration. It tells OpenCode how to work on Prometheus-MCP itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Prometheus-MCP configures →

not rated 0repo 19d ago A scan 169 tokens original MIT
View source ↗ gguatit/Prometheus-MCP
Reuse

Borrowing it

Nothing to install: this file belongs to gguatit/Prometheus-MCP. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/gguatit/Prometheus-MCP/main/.opencode/skills/prometheus-critique/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/gguatit/Prometheus-MCP

Made for: OpenCode.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for prometheus-critique

README.md
[![agentmods](https://agentmods.dev/badge/skills/gguatit/prometheus-mcp/prometheus-critique/github.svg)](https://agentmods.dev/skills/gguatit/prometheus-mcp/prometheus-critique)
Your own site
<a href="https://agentmods.dev/skills/gguatit/prometheus-mcp/prometheus-critique"><img src="https://agentmods.dev/badge/skills/gguatit/prometheus-mcp/prometheus-critique/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for prometheus-critique

Your own site · 80×15
<a href="https://agentmods.dev/skills/gguatit/prometheus-mcp/prometheus-critique"><img src="https://agentmods.dev/badge/skills/gguatit/prometheus-mcp/prometheus-critique.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 169 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,086 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00169 $0.01086
Opus 5.5 $0.00068 $0.00434
Sonnet 5.5 $0.00034 $0.00217
Haiku 4.5 $0.00017 $0.00109

Measured 19d ago against content hash 2f46710e112f, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

prometheus-critique scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 19d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.opencode/skills/prometheus-critique/SKILL.md · 46 lines

How it starts

The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Prometheus Critique Loop

Prometheus-MCP is a deterministic critic for creative artifacts. You generate and revise; Prometheus evaluates. 60 rules across 16 quality dimensions with evidence-bound scores.

When to Use

  • The user asks to review, critique, polish, or quality-check a creative artifact.
  • You just generated or modified UI / 3D scene / shader / animation / game code and want to verify quality.
  • Keywords: "이 UI 봐줘", "이거 봐줘", "검토", "크리틱", "점수", "review", "critique", "polish", "quality check".

Not for: trivial one-off snippets, or sessions where the prometheus tools are not available.

The Loop (default workflow)

  1. (Optional, grounding) If you know the matching pattern id, call get_pattern and read requiredElements / antiPatterns before generating — cheaper than fixing violations after. Use list_patterns if you don't know the id (examples: modern-landing-page, fantasy-fireball, r3f-postprocessing, top-down-shooter). Skip this whole step if the artifact already exists and no pattern applies.
  2. Generate or locate the artifact yourself (or use the user's file).
  3. Critique it: critique_artifact with { artifact: { kind, content }, patternId? } where kind is one of html | jsx | tsx | js | ts | css | json | text | markdown. For multi-file work, critique each file with its own kind (markup as html/jsx/tsx, logic modules as js/ts, styles as css); inline scripts inside HTML are covered by the html pass.
  4. Apply the top fixes yourself with your own editing tools. Read the text output's "Top fixes" plus structuredContent.suggestions / findings. Apply in priority order (P1 first); critical-severity findings are must-fix. If you'd rather get a structured plan with a ready-to-use revision prompt, call improve_artifact (plan-only by default) and apply its plan yourself.
  5. Re-critique the revised artifact. Confirm the specific findings are actually resolved and the score moved — never report an improvement you did not re-verify.
  6. Repeat until no critical/major findings remain, or the score plateaus. 2–3 rounds is normal; do not chase 10 rounds for +1 point.

Read the full file on GitHub · 46 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 19d ago First seen · 46 lines · 169 tokens per session scan A 2f46710e112f

Subscribe to this mod's changes

prometheus-critique is a skill published in the GitHub repository gguatit/Prometheus-MCP (0 stars, last pushed 19d ago), licensed MIT. It adds 169 tokens to every session and 1,086 once invoked, about $0.0007 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-18.

Creative and mediaDesign and UX
Related

prometheus-critique compared with skills from other repositories

prometheus-critique next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
prometheus-critiquegguatit/Prometheus-MCPSkill019d agoMITA169OpenCode
webgl-holographic-foilnexu-io/open-designSkill106kyesterdayApache-2.0A41Claude Code, Codex
html-ppt-hermes-cyber-terminalnexu-io/open-designSkill106kyesterdayApache-2.0A53Claude Code, Codex
html-ppt-taste-brutalistnexu-io/open-designSkill106kyesterdayApache-2.0A78Claude Code, Codex
visual-ralphYeachan-Heo/oh-my-codexSkill34k3d agoMITA52Codex
glance-canonical-layoutsandroid/skillsSkill10kyesterdayApache-2.0A56Claude Code
accessibilitymicrosoft/hve-coreSkill1.6kyesterdayMITA47Claude Code