Agent skill

Batuta Routing

by compozy in compozy/compozy

Default cost/complexity routing table for the batuta conductor.

MITAuto-check passed

Install Batuta Routing

skills CLI
$ npx skills add compozy/compozy --skill batuta-routing -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install compozy/compozy batuta-routing --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/compozy/compozy.git skills-src && mkdir -p .claude/skills && cp -r skills-src/catalog/packages/batuta/resources/skills/batuta-routing .claude/skills/batuta-routing && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
batuta-routing
GitHub stars
2.8k
Token cost
~1.1k tokens
SKILL.md length
461 words
Files
1
Skills in repo
47
Repo updated
First seen
Licence
MIT

At a glance

Default cost/complexity routing table for the batuta conductor.

  • Works in 3 steps: compozy__provider_models_list (with… → Map each lane's selection rule onto the… → Model enablement is account-side and…
  • SKILL.md covers Lane semantics (the durable…, How batuta derives the…, Canonical rule shape and Provider quirks, plus 1 more section
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Batuta Routing is an agent skill from compozy/compozy. Default cost/complexity routing table for the batuta conductor. Read at bootstrap as a starting point, validated against the live provider catalog, then stored as the per-workspace loop configuration; the stored workspace override is authoritative afterwards.

Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It works with OpenAI. The repository describes itself as: An operating system for AI agents. Plug in the agent CLIs you already use (Claude Code, Codex, Gemini CLI, Cursor) and they become a team: they split the work, hand tasks to each… The licence is MIT.

Example prompts

  • “/batuta-routing”

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. compozy__provider_models_list (with costs) is the ONLY source of
  2. Map each lane's selection rule onto the catalog using the cost fields
  3. Model enablement is account-side and invisible to the daemon — present

What it can do on your machine

Read from SKILL.md and the folder at commit eec2ed1. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are json).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Batuta Routing loads about 1.1k tokens when it runs. Until then it costs about 69 tokens; SKILL.md has 461 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~69
When it runs · the whole SKILL.md, loaded when a task matches
~1.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from compozy/compozy at commit eec2ed1, republished under its MIT licence (© compozy). 461 words, ~1,138 tokens.

Download SKILL.mdSave it as .claude/skills/batuta-routing/SKILL.md (or your agent's skills folder).
name
batuta-routing
description
Default cost/complexity routing table for the batuta conductor. Read at bootstrap as a starting point, validated against the live provider catalog, then stored as the per-workspace loop configuration; the stored workspace override is authoritative afterwards.

Batuta Routing Table

Batuta's core opinion: route every task to the cheapest executor that can handle it. Lanes use the complexity vocabulary that cy-create-tasks writes into task frontmatter (low, medium, high, critical) — the same vocabulary runtime_rules[].match.complexity matches on.

Lane semantics (the durable opinion)

LaneIntentSelection rule
lowContained change, well-trodden pathsCheapest coding-capable model in the catalog
mediumNew interfaces, moderate coordinationMid-tier coding model; raise reasoning before raising cost
highNew subsystem, heavy reasoningStrong coding model, premium tier acceptable
criticalCross-cutting, high regression riskThe operator's most trusted frontier model

How batuta derives the concrete table (never copy an example)

  1. compozy__provider_models_list (with costs) is the ONLY source of concrete provider/model IDs — it reflects the CLIs actually installed and the models actually discovered on this machine. A provider absent from the catalog is not installed; never route to it.
  2. Map each lane's selection rule onto the catalog using the cost fields (input_per_million / output_per_million) as evidence.
  3. Model enablement is account-side and invisible to the daemon — present the derived table (with costs) to the operator for confirmation before storing; ask what their accounts enable when in doubt.
Example only — derived on one machine on 2026-08-11, DO NOT reuse

On that machine the derivation produced: low → codex/gpt-5.6-luna, medium → codex/gpt-5.6-terra@high, high → codex/gpt-5.6-sol, critical → claude/claude-opus-4-8. Your catalog will differ; derive, do not copy.

Canonical rule shape

This is the exact JSON SHAPE batuta writes with compozy__loop_configure (stored per-workspace override for implement-tasks) after deriving the values from the catalog — the model/provider strings below are the same dated example as above and MUST be replaced by the derived ones. The stored override is what run-loop children resolve at execution — batuta never sends per-run rules on dispatch, because per-run rules freeze into the run and are not inherited by run-loop children anyway. Rule matching precedence inside the stored layer: id > type > complexity.

json
[
  { "match": { "complexity": "low" }, "runtime": { "provider": "codex", "model": "gpt-5.6-luna" } },
  {
    "match": { "complexity": "medium" },
    "runtime": { "provider": "codex", "model": "gpt-5.6-terra", "reasoning": "high" }
  },
  { "match": { "complexity": "high" }, "runtime": { "provider": "codex", "model": "gpt-5.6-sol" } },
  {
    "match": { "complexity": "critical" },
    "runtime": { "provider": "claude", "model": "claude-opus-4-8" }
  }
]
Show full SKILL.md (152 more words)Show less

Provider quirks

  • Some providers multiplex upstreams and require the model field to carry a prefix — e.g. opencode only binds opencode/kimi-k2.5, never bare kimi-k2.5. The catalog's exact model_id is authoritative; copy it verbatim into the rule.
  • A model can exist in the catalog and still be disabled for the operator's account at the provider (invisible to the daemon). When a lane fails its bind with zero tokens, ask the operator what their account enables.

Escalation and reclassification

  • Repeated failure in a lane: write a surgical id rule one lane up into the STORED override (compozy__loop_configure on implement-tasks, e.g. {"match":{"id":"task_NN"},"runtime":{...}} prepended to the rules), then re-dispatch batuta-deliver. id beats complexity; remove the rule after the task lands.
  • Operator reclassification in conversation ("use luna for this one") becomes the same stored id rule before the next dispatch.
  • The daemon persists resolved_runtime with per-field provenance on every generation — routing decisions are auditable via compozy__loop_status, never narrated.

© compozy, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in catalog/packages/batuta/resources/skills/batuta-routing of compozy/compozy.

Open the folder on GitHubat commit eec2ed1

Compare with similar skills

Batuta Routing next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Batuta Routing compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Batuta Routing this skillcompozy/compozy2.8k—~1.1kAutomated safety check: PassMIT
Geo Fundamentalswasp-lang/wasp19k9 repos~861Automated safety check: PassMIT
AI SDKvercel-labs/ai-facts16821 repos~1.2kAutomated safety check: PassNone
AI Image Generation and Editingzhayujie/CowAgent47k—~1.3kAutomated safety check: PassMIT
ModLens Image Vision Bridgeliustack/modlens4.2k1 repos~1.3kAutomated safety check: NotesMIT
PR Design DocOpenHands/OpenHands90k—~2.4kAutomated safety check: PassMIT

Similar skills

  • Geo Fundamentals

    wasp-lang/wasp

    Generative Engine Optimization for AI search engines (ChatGPT, Claude, Perplexity).

    19k GitHub starsUsed in 9 repos~861 tokens
    Marketing & SEOAuto-check passed
  • AI SDK

    vercel-labs/ai-facts

    Official

    Answer questions about the AI SDK and help build AI-powered features.

    168 GitHub starsUsed in 21 repos~1.2k tokens
    AI & LLM EngineeringAuto-check passed
  • Generates or edits images from text prompts through a Python script that picks an image backend based on which API keys are configured.

    47k GitHub stars~1.3k tokensUpdated today
    Media & CreativeAuto-check passed
  • Gives text-only models sight by running the modlens CLI on an image path or URL and returning structured JSON evidence with transcribed text, layout and semantics.

    4.2k GitHub starsUsed in 1 repo~1.3k tokens
    AI & LLM EngineeringAuto-check: notes
  • PR Design Doc

    OpenHands/OpenHands

    For a non-trivial pull request, write a self-contained HTML design doc under the temporary .pr/ directory and link a visibility-appropriate preview in the PR description, so maintainers grasp the…

    90k GitHub stars~2.4k tokensUpdated today
    DevelopmentAuto-check passed
  • Bibi

    JimmyLv/BibiGPT-v1

    BibiGPT CLI for summarizing videos, audio, and podcasts directly in the terminal.

    6.2k GitHub starsUsed in 1 repo~885 tokens
    Media & CreativeAuto-check passed

More from compozy/compozy

All 47 skills in this repo
  • Eng Real Scenario QA

    compozy/compozy

    Dogfoods Compozy through an autonomous startup scenario with live providers, cross-surface observation, and strict evidence audit.

    2.8k GitHub stars~3.4k tokensUpdated today
    Auto-check passed
  • Eng Test Conventions

    compozy/compozy

    Go test-shape discipline for Compozy. An agent skill from compozy/compozy.

    2.8k GitHub stars~467 tokensUpdated today
    Auto-check passed
  • Assistant UI

    compozy/compozy

    Guide for assistant-ui library - AI chat UI components. An agent skill from compozy/compozy.

    2.8k GitHub stars~958 tokensUpdated today
    Auto-check passed
  • Guide for assistant-ui UI primitives - ThreadPrimitive, ComposerPrimitive, MessagePrimitive.

    2.8k GitHub stars~999 tokensUpdated today
    Auto-check passed
  • Assistant UI Runtime

    compozy/compozy

    Guide for assistant-ui runtime system and state management. An agent skill from compozy/compozy.

    2.8k GitHub stars~856 tokensUpdated today
    Auto-check passed
  • Assistant UI Streaming

    compozy/compozy

    Guide for assistant-stream package and streaming protocols. An agent skill from compozy/compozy.

    2.8k GitHub stars~813 tokensUpdated today
    Auto-check passed

Works with

Questions about Batuta Routing

What does Batuta Routing do?

Default cost/complexity routing table for the batuta conductor. Batuta Routing is an agent skill from compozy/compozy. Default cost/complexity routing table for the batuta conductor.

How do I install Batuta Routing in Claude Code?

Run `npx skills add compozy/compozy --skill batuta-routing -a claude-code`. Or copy the skill folder (catalog/packages/batuta/resources/skills/batuta-routing in compozy/compozy) into .claude/skills/batuta-routing in your project. Claude Code loads it when a task matches its description.

How do I install Batuta Routing in Codex?

Run `npx skills add compozy/compozy --skill batuta-routing -a codex`. Or copy the skill folder (catalog/packages/batuta/resources/skills/batuta-routing in compozy/compozy) into .agents/skills/batuta-routing in your project. Codex loads it when a task matches its description.

Can I use Batuta Routing in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add compozy/compozy --skill batuta-routing -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/batuta-routing, .gemini/skills/batuta-routing, .github/skills/batuta-routing and .opencode/skills/batuta-routing in your project.

What does Batuta Routing need to run?

SKILL.md names no scripts, command-line tools or credentials: Batuta Routing is instructions for the agent only.

Does Batuta Routing access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Batuta Routing safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Batuta Routing use?

Batuta Routing is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Batuta Routing use?

About 1.1k tokens (SKILL.md is roughly 4.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Batuta Routing?

Skills that share tags, products or a category with Batuta Routing: Geo Fundamentals (wasp-lang/wasp, 19k stars), AI SDK (vercel-labs/ai-facts, 168 stars), AI Image Generation and Editing (zhayujie/CowAgent, 47k stars) and ModLens Image Vision Bridge (liustack/modlens, 4.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Batuta Routing?

compozy (a GitHub organization) maintains it in compozy/compozy, which has 2,792 GitHub stars. The repository holds 47 skills in this directory. The repository was last updated on October 8, 2026.

Source: compozy/compozy on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.