model-routing

model-routing is a cursor rule for Cursor from Oriolshhh/runware-image-mcp. It costs 1,590 tokens per session, scanned A, original, MIT.

A configuration guide for matching workflow components with model capability and reasoning effort.

In plain words
What is it for?
It is used to choose models for reviewers, builders, debuggers, councils, and other agent roles, including when to escalate or downgrade.
Why use it?
It helps teams spend more model capacity on difficult or risky work and less on bounded, mechanical tasks.

Cursor rule for Cursor

Written for Cursor: installed under .cursor/.

Good fit It is used to choose models for reviewers, builders, debuggers, councils, and other agent roles, including when to escalate or downgrade.

Compare 6 cursor rules from other repositories ↓
Install with agentmods
npx agentmods add rules/oriolshhh/runware-image-mcp/model-routing
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/Oriolshhh/runware-image-mcp

Made for: Cursor.

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 model-routing

README.md
[![agentmods](https://agentmods.dev/badge/rules/oriolshhh/runware-image-mcp/model-routing/github.svg)](https://agentmods.dev/rules/oriolshhh/runware-image-mcp/model-routing)
Your own site
<a href="https://agentmods.dev/rules/oriolshhh/runware-image-mcp/model-routing"><img src="https://agentmods.dev/badge/rules/oriolshhh/runware-image-mcp/model-routing/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 model-routing

Your own site · 80×15
<a href="https://agentmods.dev/rules/oriolshhh/runware-image-mcp/model-routing"><img src="https://agentmods.dev/badge/rules/oriolshhh/runware-image-mcp/model-routing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,590 This file is loaded in full into every session.
When invoked 1,590 The same file — it is already loaded in full.
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 $0.01590 $0.01590
Opus 5 $0.00795 $0.00795
Sonnet 5 $0.00318 $0.00318
Haiku 4.5 $0.00159 $0.00159

Measured 5d ago against content hash 93fb24ef2dfe, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

model-routing 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 5d 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.

.cursor/rules/model-routing.mdc · 77 lines

How it starts

The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Model routing

Apply the model-routing skill before delegation. Frontmatter tiers are preferred defaults, not ceilings: escalate for ambiguity, risk, irreversibility, conflicting evidence, or difficult synthesis; downgrade only after a task becomes mechanical with exact acceptance criteria. This target has no portable native per-component model field in HarnessKit. Select the closest available model manually and record any downgrade.

Component Kind Tier Effort Target selection Preload
accessibility-reviewer agent heavy high strongest available reasoning model accessible-ui-engineering, browser-visual-validation
agent-builder agent light medium fast available model model-routing
context-engineer agent standard medium balanced coding model repository-mapping, context-compression
council-chair agent heavy high strongest available reasoning model decision-synthesis
council-router agent light low fast available model specialist-routing, model-routing
debugger agent heavy high strongest available reasoning model root-cause-debugging, context-discovery
design-system-specialist agent standard medium balanced coding model design-system-integration
devils-advocate agent heavy high strongest available reasoning model tradeoff-analysis
frontend-design-lead agent heavy high strongest available reasoning model frontend-design-direction, design-system-integration
frontend-implementer agent standard medium balanced coding model frontend-guidelines, design-system-integration, ui-state-design, responsive-ui-engineering, accessible-ui-engineering
implementer agent standard medium balanced coding model minimal-diff, test-design, code-quality-guidelines
interaction-design-specialist agent standard high balanced coding model ui-state-design, motion-design, accessible-ui-engineering
motion-design-reviewer agent standard medium balanced coding model motion-design, browser-visual-validation
operations-reviewer agent standard high balanced coding model tradeoff-analysis, context-discovery
product-discovery agent standard medium balanced coding model requirements-triage, context-discovery
product-reviewer agent standard high balanced coding model tradeoff-analysis
reviewer agent heavy high strongest available reasoning model code-review, code-quality-guidelines
security-reviewer agent heavy high strongest available reasoning model threat-modeling, context-discovery
solution-architect agent heavy high strongest available reasoning model tradeoff-analysis, context-discovery
spec-engineer agent heavy high strongest available reasoning model spec-engineering, requirements-triage, specialist-routing, model-routing
task-router agent standard medium balanced coding model requirements-triage, specialist-routing, model-routing, context-discovery
test-strategist agent standard high balanced coding model test-design, context-discovery
tester agent standard medium balanced coding model test-design
ux-reviewer agent standard medium balanced coding model ui-review, ui-state-design
visual-assets-director agent heavy high strongest available reasoning model image-art-direction, visual-craft, design-system-integration, browser-visual-validation
visual-craft-specialist agent heavy high strongest available reasoning model frontend-design-direction, visual-craft, design-system-integration
visual-qa-tester agent standard medium balanced coding model browser-visual-validation
agent command standard medium balanced coding model
build-context command standard medium balanced coding model
compress-context command standard high balanced coding model
council command heavy high strongest available reasoning model
debug command heavy high strongest available reasoning model
fix command standard medium balanced coding model
frontend-design command heavy high strongest available reasoning model
implement command standard medium balanced coding model
implement-spec-frontend command heavy high strongest available reasoning model
improve-ui-ux command heavy high strongest available reasoning model
invoke command standard medium balanced coding model
pack command light low fast available model
polish-with-images command heavy high strongest available reasoning model
review command heavy high strongest available reasoning model
spec command heavy high strongest available reasoning model
spec-detailed command heavy high strongest available reasoning model
spec-frontend command heavy high strongest available reasoning model
sum-spec command light medium fast available model
ui-review command standard high balanced coding model
context-compression loop standard high balanced coding model
debug-loop loop heavy high strongest available reasoning model
frontend-implementation loop heavy high strongest available reasoning model
improve-ui-ux loop heavy high strongest available reasoning model
review-fix loop heavy high strongest available reasoning model
solution-council loop heavy high strongest available reasoning model
spec-to-implementation loop heavy high strongest available reasoning model
ui-review loop standard high balanced coding model

Read the full file on GitHub · 77 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. 5d ago First seen · 77 lines · 1,590 tokens per session scan A 93fb24ef2dfe

Subscribe to this mod's changes

model-routing is a cursor rule published in the GitHub repository Oriolshhh/runware-image-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 1,590 tokens to every session, about $0.0080 per session on Opus 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 2026-09-03.