frontier

Een lokale engine die antwoorden via één model of een panel van meerdere opdrachtregelmodellen kan laten lopen. Een beoordelaar en samensteller kunnen de uitkomsten van zo’n panel combineren.

In plain words
What is it for?
Voor het aan- of uitzetten van Frontier, het kiezen van een bekend model of preset en het samenstellen van een aangepast, alleen-lezen modelpanel.
Why use it?
Het geeft controle over wanneer gewone prompts door deze modelopstelling worden verwerkt. Je kunt de engine ook uitschakelen of eerst beschikbare modellen en instellingen bekijken.

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/mbanderas/maestro/frontier
Clone the repo
git clone --depth 1 https://github.com/mbanderas/maestro
Per session 28 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,354 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 $0.00028 $0.01354
Opus 5 $0.00014 $0.00677
Sonnet 5 $0.00006 $0.00271
Haiku 4.5 $0.00003 $0.00135

Measured 2d ago against content hash b5c394f94fb5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

frontier 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/frontier.md · 107 lines

How it starts

The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Drive the Maestro Frontier engine: a local multi-CLI fusion engine where a panel feeds a judge and synthesizer. It is opt-in: off is the default, while arming single or fusion makes ordinary later prompts auto-run through Frontier. Plugin/slash-command prompts (for example, /maestro:frontier off) bypass autorun because they are host directives.

When the user asks to compose or choose a model panel, route that request to compose; do not guess model or preset IDs. Start with catalog.

Requested action: $ARGUMENTS

Map the argument to one engine CLI call and run it. Do not edit Frontier state files by hand.

Modes

node "${CLAUDE_PLUGIN_ROOT}/bin/maestro.cjs" frontier mode off
node "${CLAUDE_PLUGIN_ROOT}/bin/maestro.cjs" frontier mode single --model <model> [--effort <level>]
node "${CLAUDE_PLUGIN_ROOT}/bin/maestro.cjs" frontier mode fusion --preset <preset> [--effort <level>]
node "${CLAUDE_PLUGIN_ROOT}/bin/maestro.cjs" frontier mode fusion --preset custom --models <model>,<model> [--effort <level>]

Use mode for a known catalog model or preset. For a custom panel, prefer compose below because it validates readiness before it arms the panel.

Catalog and composition

node "${CLAUDE_PLUGIN_ROOT}/bin/maestro.cjs" frontier catalog
node "${CLAUDE_PLUGIN_ROOT}/bin/maestro.cjs" frontier catalog --json
node "${CLAUDE_PLUGIN_ROOT}/bin/maestro.cjs" frontier compose --models <model>,<model> --dry-run
node "${CLAUDE_PLUGIN_ROOT}/bin/maestro.cjs" frontier compose --models <model>,<model> --judge <model> --synth <model> --effort <level>
node "${CLAUDE_PLUGIN_ROOT}/bin/maestro.cjs" frontier compose --models <model>,<model> --save <name>
node "${CLAUDE_PLUGIN_ROOT}/bin/maestro.cjs" frontier effort <auto|low|medium|high|xhigh|max|ultra>

frontier catalog is the source of truth for selectable models, presets, aliases, readiness, and required configuration. Do not list or invent model IDs in this command. compose accepts one to eight comma-separated models; --judge and --synth default to the first panel model. --dry-run validates without changing state; otherwise compose arms the custom fusion panel. --effort persists one validated level for all selected effort-aware Claude and Codex members. frontier effort auto returns an armed panel to provider defaults.

Read the full file on GitHub · 107 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 · 107 lines · 0 tokens per session scan A b5c394f94fb5

Subscribe to this mod's changes

frontier is a command published in the GitHub repository mbanderas/maestro (5 stars, last pushed 5d ago), licensed MIT. It adds 28 tokens to every session and 1,354 once invoked, about $0.0001 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.