Behavior Packages, end to end.

A Behavior Package is a versioned bundle that captures everything about how an LLM should behave. Instead of copying prompts between repos and wiring up evals from scratch each time, you install a Behavior Package the same way you would an npm or PyPI dependency.

A package can be made up of any combination of these typed specs:

• PromptSpec — system prompts, templates, examples
• FineTuneSpec — training recipes and configs
• EvalSpec — eval suites and benchmarks
• SafetySpec — safety policies and guardrails
• LicenseSpec — usage terms and attribution
• InstallSpec — runtime setup and dependencies
• EvidencePack — provenance and audit artifacts

A PromptSpec defines the system prompt, message templates, and example dialogues a model uses to produce consistent output. It is the smallest reusable unit on WrapperHub.

A PromptSpec includes:

• System prompt with placeholders for runtime variables
• Few-shot examples used at inference time
• Compatible models and provider-specific overrides
• Optional formatting and tool-call schemas

A FineTuneSpec is a portable training recipe. It encodes the dataset reference, hyperparameters, target base model(s), and the loss/eval signals used during training.

Because the recipe is portable, the same package can run on multiple providers (OpenAI fine-tuning API, Anthropic, your own cluster) and produce reproducible results.

An EvalSpec is a runnable bundle of evaluation cases. It captures inputs, expected behaviors, judging rubrics (heuristic or model-as-judge), and the reproducibility seed.

EvalSpec is the foundation of Green-Badge QA — a package only earns a green badge when its evals pass against the listed compatible models.

Green-Badge QA is the quality signal shown next to every package. A green badge means a package has:

• Passed every required eval at the published version
• Cleared a prompt-injection and safety scan
• A signed manifest and matching artifact hashes
• A declared, machine-readable license

If any of those checks fails, the badge degrades to review or failing, and consumers see the difference at install time.

Every Behavior Package ships with a top-level manifest.yaml. The minimal form looks like:

name: customer-support-classifier
version: 0.1.0
type: fine_tune_recipe
base_model: gpt-4.1-mini
evals:
  - label_accuracy
  - format_following
  - pii_leakage
safety:
  package_signing: required
  artifact_hashes: required
license: private-commercial

See the customer-support-classifier package for a working example.