headless-adapters

Automate headless CLI integration by mapping JSON output to evaluation schemas.

5|Updated Jan 20, 2026
One-click install
npx skills add https://github.com/youdotcom-oss/web-search-agent-evals --skill headless-adapters-youdotcom-oss
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: headless-adapters
Source: https://github.com/youdotcom-oss/web-search-agent-evals/tree/main/.plaited/skills/headless-adapters%40plaited_agent-eval-harness
Command: npx skills add https://github.com/youdotcom-oss/web-search-agent-evals --skill headless-adapters-youdotcom-oss

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Schema-driven adapters enable turning any headless CLI into a compatible, testable component for agent evaluations without writing glue code.

Core Features & Use Cases

  • Define JSON schemas to describe CLI interactions and capture structured outputs for consistent evaluation.
  • Includes guides and scaffolding to create, validate, and test adapters against CLAUDE, Gemini, Droid, and similar CLIs via the provided references.
  • Supports troubleshooting guidance to refine schemas and tool-event mappings across diverse CLI tools.

Quick Start

Define a schema for your CLI using the provided references/schema-creation-guide.md and run the headless adapter to validate it.

Frequently Asked Questions about headless-adapters

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate integration of a headless CLI for agent evaluations?

Automate headless CLI integration by mapping its JSON output to a standardized evaluation protocol. Schema-driven adapters turn any compatible CLI into a testable component for agent evaluations without writing custom glue code.

What are schema-driven adapters for headless CLI agents?

Schema-driven adapters are configurations that map structured JSON output from headless CLIs to a standardized protocol. They enable consistent testing, troubleshooting, and evaluation of agent tools like Claude Code, Gemini, and Droid.

Does this adapter approach work with Gemini and Claude Code CLIs?

Yes, the adapter approach works with Claude Code, Gemini, Droid, and similar CLIs. It requires the CLI to emit structured JSON via a supported output mode, which the schema then maps to the evaluation protocol.

How do I define a JSON schema to describe CLI interactions?

Define a JSON schema by following the provided schema-creation guide in the references directory. This scaffolding helps you capture structured outputs, validate tool-event mappings, and test the adapter against your target CLI.

Why do I need a valid SKILL.md frontmatter to use headless adapters?

A valid SKILL.md frontmatter with name and description is required to establish the adapter's metadata. This ensures the headless adapter can correctly process the CLI's structured JSON output within the standardized evaluation protocol.

What are the limitations of using schema-driven adapters for headless CLIs?

Schema-driven adapters require the headless CLI to emit structured JSON via a supported output mode. CLIs lacking structured JSON output cannot be mapped to the standardized evaluation protocol without additional custom preprocessing.