v-models

A command for updating the model choices used by Compound V, a system that sends tasks to different coding backends. It discovers available models and maps them to the labels deep, standard, or light.

In plain words
What is it for?
Use it to refresh one backend or all supported backends, review the discovered choices, assign tiers, and save the routing map in .claude/compound-v.json.
Why use it?
It avoids having to edit model names throughout a project when providers change or replace their models.

Command

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 commands/procoders/superpowers-v/v-models
Clone the repo
git clone --depth 1 https://github.com/procoders/superpowers-v
Per session 73 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,125 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 $0.00073 $0.05125
Opus 5 $0.00036 $0.02563
Sonnet 5 $0.00015 $0.01025
Haiku 4.5 $0.00007 $0.00513

Measured 2d ago against content hash 7901997e3e56, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

v-models 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 2d 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.

commands/v-models.md · 345 lines

How it starts

The opening of the file, as written. The whole thing — 345 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are running /v:models — the Compound V model broker's refresh surface. Compound V routes work by intent (a stable tier vocabulary — deep / standard / light) instead of hardcoding model strings that rot every time a provider ships a new model. The mapping from tier → concrete model lives in a refreshable models block in .claude/compound-v.json. This command discovers what each backend can run right now, lets you assign each tier, and rewrites that block. Nothing else in the plugin changes — the dispatcher resolves tiers through scripts/compound-v-resolve-model.py at dispatch time, so refreshing the map here is the only thing you ever touch when models churn.

Argument (optional): {{args}} may name a single backend to refresh in isolation (claude | codex | antigravity | cursor | devin | opencode); otherwise walk all of them. devin and opencode are worker-only backends (v1) — their model maps drive dispatch only, never any arbiter/review panel seat.

This is the "skill picks the models and offers you the options" surface. Do the discovery, show what you found, then let the user choose. Never silently pick a model the user did not confirm. NEVER assign haiku to any tier on any backend.


Step 0 — Load the current map

Read .claude/compound-v.json if it exists. Remember its current models block (seeded by /v:init) so you can show the user what is changing and preserve any backend they don't refresh this run. The models block is per-stance — shape {<stance>: {<backend>: {<tier>: model}}}. If the file or its models key is absent, fall back to the built-in default (the resolver carries the same one). Only the claude rows differ across stances — cost-aware.claude.standard is sonnet, everywhere else standard Claude is opus:

"models": {
  "balanced": {
    "claude":      { "deep": "opus",                  "standard": "opus",                  "light": "sonnet" },
    "codex":       { "deep": "gpt-5.6-sol",            "standard": "gpt-5.6-terra",          "light": "gpt-5.6-luna" },
    "antigravity": { "deep": "Gemini 3.1 Pro (High)", "standard": "Gemini 3.1 Pro (Low)", "light": "Gemini 3.5 Flash (Low)" },
    "cursor":      { "deep": "auto",                  "standard": "auto",                  "light": "auto" }
  },
  "cost-aware": {
    "claude":      { "deep": "opus",                  "standard": "sonnet",                "light": "sonnet" },
    "codex":       { "deep": "gpt-5.6-sol",            "standard": "gpt-5.6-terra",          "light": "gpt-5.6-luna" },
    "antigravity": { "deep": "Gemini 3.1 Pro (High)", "standard": "Gemini 3.1 Pro (Low)", "light": "Gemini 3.5 Flash (Low)" },
    "cursor":      { "deep": "auto",                  "standard": "auto",                  "light": "auto" }
  }
  // conservative + claude-only mirror balanced
}

/v:models writes this per-stance shape; the resolver still accepts the legacy flat shape {<backend>: {<tier>: model}} (applied to every stance) for backward-compat. If {{args}} named one backend, only discover + reassign that backend (across every stance's block) and leave the other backends exactly as they are.


Read the full file on GitHub · 345 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. 2d ago First seen · 345 lines · 73 tokens per session scan A 7901997e3e56

Subscribe to this mod's changes

v-models is a command published in the GitHub repository procoders/superpowers-v (35 stars, last pushed 2d ago), licensed MIT. It adds 73 tokens to every session and 5,125 once invoked, about $0.0004 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 2026-08-30.