setup_routing

A setup guide for routing different project tasks to different AI models and reasoning levels. It separates tasks needing deeper analysis from high-volume mechanical work and configures rules for both.

In plain words
What is it for?
Use it to configure per-project model and effort rules for coding, review, debugging, and other development phases.
Why use it?
It helps control model usage by matching the level of reasoning to the task instead of using the same setting for everything.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/blendsdk/claude-codeops/setup_routing
Any agent
npx skills add blendsdk/claude-codeops --skill setup_routing
Clone the repo
git clone --depth 1 https://github.com/blendsdk/claude-codeops

Made for: Claude Code, Codex.

Per session 180 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,589 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00180 $0.03589
Opus 5 $0.00090 $0.01795
Sonnet 5 $0.00036 $0.00718
Haiku 4.5 $0.00018 $0.00359

Measured yesterday against content hash 5590bb88458d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

setup_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 yesterday.

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.

skills/setup_routing/SKILL.md · 220 lines

How it starts

The opening of the file, as written. The whole thing — 220 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Model & Effort Routing Setup (setup_routing)

CodeOps Skills Version: 3.20.0

Configure per-project model and effort routing for the project the user is currently in, so that expensive reasoning (Opus, high/xhigh thinking) is spent only where it changes output quality, and high-volume mechanical work runs on Sonnet. Invoked as /codeops:setup_routing or the typeable alias /setup_routing.

The two-layer principle (preserve this throughout)

Routing has two layers, and they must stay coordinated:

  1. Policy layer (soft, behavioral). A block in the project's CLAUDE.md decides which model each phase runs on — inline-first — and which pinned executor backs a dispatch when one is warranted, expressed as a rule over task tags.
  2. Enforcement layer (hard, guaranteed). Each executor subagent's frontmatter pins both model: (which model runs) and effort: (which reasoning effort runs) when that executor is invoked. Claude Code enforces both, overriding the user's session/settings.json/env config — so routing works regardless of what the user has already configured. The sole exception is the CLAUDE_CODE_SUBAGENT_MODEL env var, which sits above subagent frontmatter and forces every subagent onto one model (a deliberate global cost-cap escape hatch — surface it in Phase 5).

The policy is only trustworthy because the pinned executors exist to back it. Since v3.2.0 the two executors (plan-task-executor, plan-task-executor-opus) ship with the plugin in its agents/ directory — every install has them, so the policy block is valid on its own. Never generate routing prose that references an executor that exists neither in the plugin nor in the project. Per-project copies under .claude/agents/ are an opt-in override for users who want customized executor prompts (a project agent of the same name shadows the plugin one).

The policy layer is a behavioral instruction to the orchestrator, not a hard guarantee. The exec_plan skill's "Execution mode — inline first" protocol (execution-protocol.md) defines when a phase may be dispatched at all, the phase packet, the parent/executor division of labor, and a missing-executor guard (inline fallback with notice). Watching one real phase run under the policy is still the recommended validation (Phase 5).

Read the full file on GitHub · 220 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. yesterday First seen · 220 lines · 180 tokens per session scan A 5590bb88458d

Subscribe to this mod's changes

setup_routing is a skill published in the GitHub repository blendsdk/claude-codeops (4 stars, last pushed 1mo ago), licensed MIT. It adds 180 tokens to every session and 3,589 once invoked, about $0.0009 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 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens