Index of every page on this site: llms.txt
All skills
a2ui-project avatar

/inference-format-optimizer

@6bf4cc2
by A2UIa2ui-project/a2ui18k stars
1,430
Run it yourself

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.

  • 19 files
  • 207.9 KB
  • Updated 2 weeks ago
  • GitHub
Use this Skill: https://skilld.dev/gh/a2ui-project/a2ui/inference-format-optimizer

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

referencesscoring_model.md

≈499 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Formal Scoring Model & Decision Guardrails

This reference details the quantitative decision model, correctness guardrails, efficiency caps, and composite score formula ($S_{\text{opt}}$) for A2UI format optimizations.


1. Correctness Guardrails (Non-Negotiable)

Every candidate format iteration must pass all correctness guardrails. If any guardrail fails, the iteration MUST BE REVERTED:

  1. Pytest Unit Conformance: Must be PASS (100% unit tests passing).
  2. Algorithmic Schema Pass Rate (SchemaAcc): Pass rate of output payloads against the target catalog JSON schema (a2ui_scorer). Must be $\ge$ Baseline.
  3. Quality Score (QualityScore): Model-graded QA semantic intent match (measured_model_graded_qa). Must be $\ge$ Baseline.

2. Efficiency Regression Caps (Non-Negotiable REVERT Triggers)

An iteration MUST BE REVERTED if any of the following efficiency caps are exceeded relative to baseline/previous run:

  • Code Output Tokens: Increases by > 5% (prevents format verbosity expansion).
  • Streaming Latency (Non-reasoning Output Time): Increases by > 10% (prevents code streaming bottlenecks).
  • Reasoning Tokens: Increases by > 15% (prevents prompt instruction search space ambiguity).

3. Composite Optimization Score ($S_{\text{opt}}$)

The composite score $S_{\text{opt}}$ balances accuracy gains against token and latency overheads:

[ S_{\text{opt}} = 0.50 \cdot \text{SchemaAcc} + 0.30 \cdot \text{QualityScore} - 0.15 \cdot \left(\frac{\text{CodeTok}}{\text{BaseCodeTok}}\right) - 0.05 \cdot \left(\frac{\text{ReasonTok}}{\text{BaseReasonTok}}\right) - 0.03 \cdot \left(\frac{\text{InputTok}}{\text{BaseInputTok}}\right) ]

Decision Rule:

  • If $S_{\text{opt}}(\text{Current}) > S_{\text{opt}}(\text{Baseline})$ $\rightarrow$ KEEP CHANGE (--status KEEP)
  • If $S_{\text{opt}}(\text{Current}) \le S_{\text{opt}}(\text{Baseline})$ $\rightarrow$ REVERT CHANGE (--status REVERT)

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at 6bf4cc2. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 2 weeks ago
metadata
{
  "internal": true
}

README badge

README badge for a2ui-project/a2ui/inference-format-optimizer