skills.rest
SynthFlowAI
SynthFlowAIOFFICIAL
Explore all skills by this author →
WebsiteGitHub

create-eval

Create custom evaluations for Synthflow AI agents with CUSTOM_EVAL actions.

Updated Mar 23, 2026
Search Tags:#metrics#score#api-key#assistant#synthflow#custom-eval#post-call
One-click install
npx skills add https://github.com/SynthFlowAI/synthflow-skills --skill create-eval-synthflowai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: create-eval
Source: https://github.com/SynthFlowAI/synthflow-skills/tree/main/create-eval
Command: npx skills add https://github.com/SynthFlowAI/synthflow-skills --skill create-eval-synthflowai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automate the creation and integration of post-call evaluations to consistently score agent performance and identify improvement areas.

Core Features & Use Cases

  • Define CUSTOM_EVAL actions with identifiers, prompts, and scoring categories to tailor evaluations for your workflows.
  • Attach evaluations to assistants to automatically run after calls and generate structured results.
  • Use for building scorecards, QA dashboards, and continuous improvement programs.

Quick Start

Add a CUSTOM_EVAL action to an assistant with a unique identifier, prompt text, and category, then configure your API key.

Frequently Asked Questions about create-eval

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

FAQPage Schema
How do I set up custom post-call evaluations for AI voice agents?▼

Configuring post-call analysis requires adding a CUSTOM_EVAL action with a question object containing identifier, text, category, and expected_result. This automatically scores call outcomes for your assistants.

What is a custom evaluator scorecard for call quality tracking?▼

A custom evaluator scorecard uses CUSTOM_EVAL actions to define success metrics, prompts, and scoring categories that quantify agent performance. This automates call quality measurement and identifies improvement areas.

Do I need a Synthflow API key to create custom evaluations?▼

Yes, creating custom evaluations requires a Synthflow API key (SYNTHFLOW_API_KEY) and internet access. The API key authenticates requests to attach CUSTOM_EVAL actions to your assistants.

Can I attach automated scoring to my existing Synthflow AI assistants?▼

Yes, you can attach CUSTOM_EVAL actions to existing assistants to automatically run after calls and generate structured scoring results, enabling automated QA dashboards for your workflows.

How do I define success metrics for enterprise call scenarios?▼

Define success metrics by configuring a CUSTOM_EVAL action with a category and expected_result for enterprise call scenarios. This tailors post-call analysis to track targeted outcomes and score agent performance.

Related Skills

Explore More Data & Analytics Skills→
SynthFlowAISynthFlowAI

synthflow-voice-ai

Automate AI voice agent creation and management via the Synthflow API.

Official
Advanced
neyrojasjneyrojasj

evaluator

Evaluate code against QA checklists and emit PASS/FAIL verdicts.

Community
Advanced
SynthFlowAISynthFlowAI

manage-actions

Configure transfers, SMS, webhooks, and evaluations for Synthflow assistants.

Official
Advanced
sih3rronsih3rron

gong-call-intelligence

Search Gong call transcripts and extract structured highlights via gtmService tools.

Community
Intermediate