New: 97,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Inference Format Optimizer

by a2ui-project in a2ui-project/a2ui

Iterative benchmarking, evaluation, and algorithmic optimization of alternative A2UI inference formats (such as Express, Atom, and Elemental).

Apache-2.0Auto-check passedDevelopment

Install Inference Format Optimizer

skills CLI
$ npx skills add a2ui-project/a2ui --skill inference-format-optimizer -a claude-code

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

GitHub CLI
$ gh skill install a2ui-project/a2ui inference-format-optimizer --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/a2ui-project/a2ui.git skills-src && mkdir -p .claude/skills && cp -r skills-src/eval/iterative_format_optimizer/skills/inference-format-optimizer .claude/skills/inference-format-optimizer && 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
inference-format-optimizer
GitHub stars
17k
Token cost
~985 tokens
SKILL.md length
227 words
Files
19 (incl. scripts, references)
Skills in repo
20
Repo updated
First seen
Licence
Apache-2.0

At a glance

Iterative benchmarking, evaluation, and algorithmic optimization of alternative A2UI inference formats (such as Express, Atom, and Elemental).

  • Works in 6 steps: Analyze History: Inspect past runs in… → Implement Hypothesis: Modify… → Run Unit Conformance Tests: Verify code… → …
  • Run optimization passes
  • SKILL.md covers Quick-Start CLI Cheatsheet, Detailed References and The 6-Step Optimization Workflow
  • Runs Python scripts from its folder; calls python and git

What it does

Inference Format Optimizer is an agent skill from a2ui-project/a2ui. Iterative benchmarking, evaluation, and algorithmic optimization of alternative A2UI inference formats (such as Express, Atom, and Elemental). Trigger when asked to: (1) Run optimization passes or loops on an inference format, (2) Evaluate or benchmark format accuracy, latency, or token efficiency, (3) Create parallel worktree subagents for format iteration, or (4) Benchmark format trade-offs against baselines.

Its SKILL.md is about 990 tokens, which your agent loads only when the skill is triggered. The skill folder holds 23 other files, including scripts and reference files (for example `references/agent_instructions.md`, `references/inference_format_iteration.md` and `references/scoring_model.md`).

It sits in Development, covering Git worktrees and Subagents. It works with Python. The licence is Apache-2.0.

When your agent uses it

  • Run optimization passes
  • Loops on an inference format
  • Benchmark format accuracy
  • Token efficiency

Example prompts

  • “/inference-format-optimizer”

Requirements

  • Python 3

Workflow steps

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

  1. Analyze History: Inspect past runs in eval/iterative_format_optimizer/history// and read…
  2. Implement Hypothesis: Modify compiler.py, prompt_generator.py, or parser.py under python/a2ui_agent/src/a2ui/inference_formats/experimental…
  3. Run Unit Conformance Tests: Verify code changes pass pytest unit tests.
  4. Execute Benchmark Evaluation: Run python scripts/optimize_format.py --format .
  5. Evaluate Decision Rules
  6. Archive & Synchronize: Archive run with --archive and update history index using python scripts/sync_history.py.

What it can do on your machine

Read from SKILL.md and the folder at commit ae466ff. 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

    Ships 9 files in scripts/ (Python, from the files we listed), which the agent can run.

    Shell commands in SKILL.md call:

    • python
    • git

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

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    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

Inference Format Optimizer loads about 985 tokens when it runs, and up to ~8.1k if it reads all its reference files. Until then it costs about 110 tokens; SKILL.md has 227 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~110
When it runs · the whole SKILL.md, loaded when a task matches
~985
With references · SKILL.md plus every file in references/, read only if the agent opens them
~8.1k

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); the scripts in this folder are not scanned.

SKILL.md

The full file from a2ui-project/a2ui at commit ae466ff, republished under its Apache-2.0 licence (© a2ui-project). 227 words, ~985 tokens.

Download SKILL.mdSave it as .claude/skills/inference-format-optimizer/SKILL.md (or your agent's skills folder). This skill also uses 18 other files; get the full folder from GitHub.
name
inference-format-optimizer
description
Iterative benchmarking, evaluation, and algorithmic optimization of alternative A2UI inference formats (such as Express, Atom, and Elemental). Trigger when asked to: (1) Run optimization passes or loops on an inference format, (2) Evaluate or benchmark format accuracy, latency, or token efficiency, (3) Create parallel worktree subagents for format iteration, or (4) Benchmark format trade-offs against baselines.

Inference Format Optimizer

This skill provides procedural workflows, CLI orchestrators, decision guardrails, and subagent protocols for iteratively optimizing A2UI inference formats.

Quick-Start CLI Cheatsheet

All execution scripts live under scripts/ in this skill:

ActionExecutable Command
Run Fast Validation Evalpython scripts/optimize_format.py --format <format>
Run Full Evaluation Suitepython scripts/optimize_format.py --format <format> --full
Test Parsing / Compilationpython scripts/optimize_format.py --format <format> --compile "(Card (Text \"Hi\"))"
Compare vs Baselinepython scripts/compare_results.py --baseline eval/iterative_format_optimizer/baselines/<format>/unbounded_run_meta.json eval/iterative_format_optimizer/logs/temp_optimization/
Archive Run Artifactspython scripts/optimize_format.py --format <format> --archive --hypothesis "..." --status KEEP [--history-dir <path>]
Sync Multi-Worktree Historypython scripts/sync_history.py [--history-dir <path>]

Detailed References

The 6-Step Optimization Workflow

  1. Analyze History: Inspect past runs in eval/iterative_format_optimizer/history/<format>/ and read eval/iterative_format_optimizer/history_summary.md to avoid repeating past reverted hypotheses.
  2. Implement Hypothesis: Modify compiler.py, prompt_generator.py, or parser.py under python/a2ui_agent/src/a2ui/inference_formats/experimental/<format>/.
  3. Run Unit Conformance Tests: Verify code changes pass pytest unit tests.
  4. Execute Benchmark Evaluation: Run python scripts/optimize_format.py --format <format>.
  5. Evaluate Decision Rules:
    • Must pass Pytest and maintain baseline accuracy.
    • Code Output Tokens must NOT expand $> +5%$.
    • Keep change if composite score $S_{\text{opt}}$ improves; revert otherwise (git reset --hard HEAD).
  6. Archive & Synchronize: Archive run with --archive and update history index using python scripts/sync_history.py.

© a2ui-project, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 18 other files (scripts, references) in eval/iterative_format_optimizer/skills/inference-format-optimizer of a2ui-project/a2ui.

  • SKILL.md
  • references/agent_instructions.md
  • references/inference_format_iteration.md
  • references/scoring_model.md
  • references/subagent_protocol.md
  • scripts/README.md
  • scripts/compare_results.py
  • scripts/optimize_format.py
  • scripts/sync_history.py
  • scripts/utils/__init__.py
  • scripts/utils/archiver.py
  • scripts/utils/format_tools.py
  • scripts/utils/reporter.py
  • scripts/utils/runner.py
  • templates/subagent_prompt.md
  • tests/test_compare_results.py
  • … and 3 more

Open the folder on GitHubat commit ae466ff

Compare with similar skills

Inference Format Optimizer 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.

Inference Format Optimizer compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Inference Format Optimizer this skilla2ui-project/a2ui17k—~985Automated safety check: PassApache-2.0
Worktree Env Setupmeta-pytorch/attention-gym1.3k—~858Automated safety check: PassBSD-3-Clause
Release Candidate Prepopenai/openai-agents-python30k—~4.9kAutomated safety check: PassMIT
Cursor Composer Task DelegateChachamaru127/claude-code-harness3.2k—~4.4kAutomated safety check: NotesMIT
Bench Batonnooga/let-go568—~821Automated safety check: PassMIT
Burla Parallel Dev ClustersBurla-Cloud/burla263—~1.6kAutomated safety check: PassCustom licence

Similar skills

  • Worktree Env Setup

    meta-pytorch/attention-gym

    Sets up an isolated per-worktree Python environment for attention-gym development using nightly PyTorch and the CI-mirroring uv flow.

    1.3k GitHub stars~858 tokensUpdated yesterday
    DevelopmentAuto-check passed

  • Release Candidate Prep

    openai/openai-agents-python

    Official

    Prepare a local Python SDK release candidate in a dedicated worktree.

    30k GitHub stars~4.9k tokensUpdated today
    DevelopmentAuto-check passed

  • Cursor Composer Task Delegate

    Chachamaru127/claude-code-harness

    Hands one implementation task to Cursor Composer in an isolated git worktree, then reviews its diff and cherry-picks the result into the main branch.

    3.2k GitHub stars~4.4k tokensUpdated 2 days ago
    DevelopmentAuto-check: notes

  • Bench Baton

    nooga/let-go

    Coordinate heavy local workloads across worktrees, processes, and subagents — benchmarks and timing-sensitive gates run exclusively on a quiesced machine, while builds, test suites, regeneration…

    568 GitHub stars~821 tokensUpdated today
    DevelopmentAuto-check passed

  • Burla Parallel Dev Clusters

    Burla-Cloud/burla

    Sets up an isolated Burla dev cluster per git worktree so several agents can work in parallel, and explains when to use local-dev or remote-dev.

    263 GitHub stars~1.6k tokensUpdated 15 days ago
    DevelopmentAuto-check passed

  • Spec Kitty Git Workflow

    spec-kitty/spec-kitty

    Understand how Spec Kitty manages git: what git operations Python handles automatically, what agents must do manually, worktree lifecycle, auto-commit behavior, merge execution, and the safe-commit…

    1.7k GitHub stars~2.3k tokensUpdated today
    DevelopmentAuto-check passed

More from a2ui-project/a2ui

All 20 skills in this repo

  • A2ui Remediate Problem

    a2ui-project/a2ui

    Remediates a specific recommendation from an A2UI compliance report issue, or resolves any general GitHub issue, by inspecting context, implementing minimal targeted fixes, verifying tests, creating…

    17k GitHub stars~1.5k tokensUpdated today
    Auto-check passed

  • A2ui Issue Triage

    a2ui-project/a2ui

    Automates the triage of GitHub issues in the A2UI repository.

    17k GitHub stars~1.5k tokensUpdated today
    Auto-check passed

  • A2ui Audit

    a2ui-project/a2ui

    Main coordination skill to run the blueprint compliance, documentation synchronization, and test quality audits, posting the combined results as a labeled GitHub issue.

    17k GitHub stars~1.6k tokensUpdated today
    Auto-check passed

  • A2ui Generate Pydantic Models

    a2ui-project/a2ui

    Automated generator for strongly typed Pydantic v2 data models and basic catalogs across any A2UI protocol version (v0.8, v0.9, v0.9.1, v1.0, etc.).

    17k GitHub stars~2.3k tokensUpdated today
    Auto-check passed

  • A2ui Python Development

    a2ui-project/a2ui

    Grounding, architectural standards, Python best practices, package facade conventions, testing, and verification workflows for developing any Python code across the entire A2UI repository (core…

    17k GitHub stars~2.4k tokensUpdated today
    Auto-check passed

  • A2ui Release Python

    a2ui-project/a2ui

    Releases the Python SDKs (a2ui-core, a2ui-agent-sdk) to PyPI by running preflight checks, dispatching the release workflow as a dry run, confirming with the maintainer, publishing, and following the…

    17k GitHub stars~4.2k tokensUpdated today
    Auto-check passed

Works with

Categories

Questions about Inference Format Optimizer

What does Inference Format Optimizer do?

Iterative benchmarking, evaluation, and algorithmic optimization of alternative A2UI inference formats (such as Express, Atom, and Elemental). Inference Format Optimizer is an agent skill from a2ui-project/a2ui. Iterative benchmarking, evaluation, and algorithmic optimization of alternative A2UI inference formats (such as Express, Atom, and Elemental).

When should I use Inference Format Optimizer?

Inference Format Optimizer fits situations like: run optimization passes; loops on an inference format; benchmark format accuracy; token efficiency.

How do I install Inference Format Optimizer in Claude Code?

Run `npx skills add a2ui-project/a2ui --skill inference-format-optimizer -a claude-code`. Or copy the skill folder (eval/iterative_format_optimizer/skills/inference-format-optimizer in a2ui-project/a2ui) into .claude/skills/inference-format-optimizer in your project. Claude Code loads it when a task matches its description.

How do I install Inference Format Optimizer in Codex?

Run `npx skills add a2ui-project/a2ui --skill inference-format-optimizer -a codex`. Or copy the skill folder (eval/iterative_format_optimizer/skills/inference-format-optimizer in a2ui-project/a2ui) into .agents/skills/inference-format-optimizer in your project. Codex loads it when a task matches its description.

Can I use Inference Format Optimizer 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 a2ui-project/a2ui --skill inference-format-optimizer -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/inference-format-optimizer, .gemini/skills/inference-format-optimizer, .github/skills/inference-format-optimizer and .opencode/skills/inference-format-optimizer in your project.

What does Inference Format Optimizer need to run?

Going by SKILL.md and its folder, Inference Format Optimizer needs Python for the scripts in its folder and the command-line tools its instructions call (python and git). Our summary lists: Python 3.

Does Inference Format Optimizer access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Inference Format Optimizer 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. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Inference Format Optimizer use?

Inference Format Optimizer is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Inference Format Optimizer use?

About 985 tokens (SKILL.md is roughly 3.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 7.1k tokens, read only when the agent opens those files.

What are the alternatives to Inference Format Optimizer?

Skills that share tags, products or a category with Inference Format Optimizer: Worktree Env Setup (meta-pytorch/attention-gym, 1.3k stars), Release Candidate Prep (openai/openai-agents-python, 30k stars), Cursor Composer Task Delegate (Chachamaru127/claude-code-harness, 3.2k stars) and Bench Baton (nooga/let-go, 568 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Inference Format Optimizer?

a2ui-project (a GitHub organization) maintains it in a2ui-project/a2ui, which has 16,602 GitHub stars. The repository holds 20 skills in this directory. The repository was last updated on October 7, 2026.

Source: a2ui-project/a2ui on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.