Create Adapter

Create Adapter

Bootstrap a new benchmark adapter in the Harbor repository. This skill scaffolds the adapter directory with harbor adapter init and then defers to the adapter tutorial for every implementation decision.

Authoritative reference

The adapter tutorial is the authoritative specification for this skill. Read it in full before taking any action beyond scaffolding:

docs/content/docs/datasets/adapters.mdx

That path is relative to the Harbor repo root (a skill prerequisite — see below). The tutorial contains:

• Required directory structures for the generated tasks and the adapter code package.
• Step-by-step process (steps 1-9) covering benchmark analysis, oracle verification, parity experiments, dataset registration, and submission.
• Schemas for task.toml, parity_experiment.json, adapter_metadata.json, and dataset.toml.
• Parity matching criterion, pre-flight checklist, and debug playbook.
• README format rules (machine-parsed; deviations break automation).

Do not substitute prior knowledge for the contents of that file. Treat it as the contract.

Prerequisites

• Harbor CLI is installed and available on PATH (harbor --version succeeds).
• Working directory is the Harbor repository root.
• Harbor checkout is current: run git fetch origin && git status and pull main if behind. Stale checkouts miss recent adapter and agent fixes and are a common source of spurious parity failures later on.

Workflow

1. Read the tutorial

Before any filesystem or CLI action, Read docs/content/docs/datasets/adapters.mdx in full. Pay particular attention to:

• Required Directory Structures — the contract for generated task and adapter layouts.
• Step 1. Understand the Original Benchmark — what to identify upstream before coding.
• Step 8. Register the Dataset → Naming rules — the name field and <org>/<task> format requirements.

2. Gather benchmark context from the user

Collect the following before scaffolding. If the user has not provided an item, ask before proceeding — these inputs shape the scaffold and the tutorial steps that follow.

3. Create the adapter branch

Per tutorial Step 2, work on a dedicated branch:

git checkout -b <adapter-name>-adapter

4. Run the scaffold

Prefer the non-interactive form when both names are known:

harbor adapter init <adapter-name> --name "<Human-Readable Name>"

Otherwise run harbor adapter init interactively and let the CLI prompt.

Expected output: a new directory at adapters/<adapter-name>/ containing pyproject.toml, README.md, src/<adapter_name>/, and the template task files under src/<adapter_name>/task-template/. Verify the directory exists before continuing.

5. Hand off to the tutorial

Continue from "Step 1. Understand the Original Benchmark" in the tutorial. Do not invent structure, field names, or workflow beyond what the guide specifies.

High-priority gotchas (each is documented in the tutorial, but these are the most common adapter-build failures — surface them proactively as you work through the steps):

• Every generated task.toml must contain a name field under [task]. main.py is responsible for deriving a sanitized, unique, registry-safe name for every task. Tasks without a name cannot be registered. See the tutorial's "Naming rules" table.
• Task names must be stable across adapter runs. Unstable names churn registry digests on republish. If upstream lacks stable identifiers, mint a deterministic scheme (e.g., {dataset}-1, {dataset}-2) from a reproducible sort.
• version = "1.0" in task.toml is the schema version — leave it alone. Dataset versions are publish-time tags requested in the PR description, not a field in task.toml or dataset.toml.
• main.py must support --output-dir, --limit, --overwrite, and --task-ids. These flags are required for reproducible runs and task-level debugging.
• The generated README.md is parsed by downstream automation. Fill in every section exactly as the template defines; put extra context in the Notes section or in the notes fields of parity_experiment.json / adapter_metadata.json. Do not add, rename, reorder, or remove sections.
• Do not run parity experiments unilaterally. Tutorial Step 4 requires team coordination on agents, models, and number of runs before incurring API costs. Complete sanity checks first, and execute full runs symmetrically on both sides.

Reference adapters by scenario

When implementation questions come up, point at an existing adapter that matches the benchmark's shape:

What this skill does NOT do

• Implement adapter.py, main.py, or any task-template files. Those are the contributor's work, guided by the tutorial.
• Run oracle verification or parity experiments. Those require explicit user intent and, for parity, team coordination.
• Publish the dataset or open PRs. Those are later steps in the tutorial.

Failure modes