Evaluation Debugging

Organization

VectorSpaceLab

evaluation-debugging

Design and run ADK evaluation, test, and debugging workflows for eval sets, JSON fixtures, event traces, sessions, and safe native verification.

Overview

Publisher	VectorSpaceLab
Repository	`AREX-Skill`
Skill name	`evaluation-debugging`
Stars	341
Forks	26
Bundled files	4
License	Apache-2.0
Links	Skill source Repository

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
4 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by VectorSpaceLab on GitHub. Read the source before you install it.

Installation

Install the Evaluation Debugging AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/adk-python/sub-skills/evaluation-debugging
Enable the skill in any chat where you want it available.

Store

Installed

Skills

Add skills

Import one or more skills from a public GitHub folder or a local .skill or .zip archive, or SKILL.md file.

From GitHub URLFrom .skill, .zip, or SKILL.md

GitHub repo or folder

https://github.com/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/adk-python/sub-skills/evaluation-debugging

Points to a single skill folder, or any folder that contains multiple skills. Public repos only.

CancelContinue

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/VectorSpaceLab/AREX-Skill.git /tmp/AREX-Skill
mkdir -p .claude/skills
cp -r /tmp/AREX-Skill/skills/repositories/repo-skills/adk-python/sub-skills/evaluation-debugging .claude/skills/evaluation-debugging

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Evaluation Debugging in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Evaluation Debugging on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Evaluation Debugging is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Evaluation Debugging

Use this sub-skill when the user needs to create or run ADK evaluation cases, replay JSON tests, inspect session events, summarize traces, or choose safe native verification targets for an ADK agent.

Route Common Requests

Convert a manual interaction into an adk test fixture or replay existing tests/*.json event files with deterministic mocks: use Evaluation Workflows.
Run metric-based EvalSet checks with adk eval, adk eval_set, or google.adk.evaluation.AgentEvaluator: use Evaluation Workflows.
Debug missing tool calls, unexpected output, callback order, branch/node routing, or event/session state: use Debugging.
Diagnose schema validation, credentials, flaky LLM metrics, missing trace attributes, and server session lookup failures: use Troubleshooting.
Summarize raw ADK session/event JSON without exposing long values: run summarize_adk_events.py.

Boundaries

Stay in this sub-skill for adk test, adk eval, adk eval_set, JSON test/eval fixtures, AgentEvaluator, event printing, session/trace inspection, and selecting safe native eval/debug verification candidates.
Route general CLI command discovery, agent app layout, server startup, deployment, and YAML configuration mechanics to cli-configuration-deployment.
Route code changes to agent constructors, tools, callbacks, instructions, schemas, or model binding to agent-construction or tools-and-integrations.
Route repository maintainer style, broad pytest policy, docs/sample update policy, and PR review gates to repo-development.

Safe Defaults

Prefer read-only checks first: adk test --help, adk eval --help, adk eval_set --help, fixture validation, event summaries, and focused native candidates.
Do not start a long-running adk web or api_server process unless the user asks or confirms; if one is already running, inspect it before starting another.
Treat model credentials, cloud storage, Vertex AI scenario generation, and LLM-as-judge metrics as environment-dependent; separate these from deterministic local replay.
Never print full tool arguments, session state, trace payloads, or model requests when they may contain secrets; truncate and redact before sharing.

Quick Decision Guide

Goal	Best path	Why
Regression-test exact event flow	`adk test AGENTS_DIR` with `tests/*.json`	Replays stored event JSON, mocks model outputs/randomness, and normalizes volatile IDs.
Score behavior against references/tools	`adk eval AGENT_INIT EVAL_SET` or `AgentEvaluator.evaluate(...)`	Uses EvalSet metrics such as `tool_trajectory_avg_score` and `response_match_score`.
Inspect why a web session failed	Fetch session JSON, summarize events, then inspect `/dev/apps/{app}/debug/trace/session/{session}`	Correlates events, node paths, tool calls, and LLM/tool spans.
Choose repo-native verification	Start with focused evaluation unit coverage, selected eval fixture integration coverage, and CLI help	Avoids broad, credential-heavy, or server-dependent runs unless needed.

Bundled Reference Map

Evaluation Workflows — EvalSet schemas, legacy JSON, CLI/API routes, fixture design, assertions, and native candidates.
Debugging — adk run/web debugging, session and trace inspection, event fields, callback order, and print_event usage.
Troubleshooting — Concrete failure signatures for schema mismatches, credentials, flaky metrics, missing tools, sessions, and trace attributes.
summarize_adk_events.py — Safe stdin/file event JSON summarizer for sessions, event arrays, and trace-like dumps.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Evaluation Debugging AI skill do?

Design and run ADK evaluation, test, and debugging workflows for eval sets, JSON fixtures, event traces, sessions, and safe native verification.

Why use Evaluation Debugging on TypingMind?

Because you install it once and use it with any model. Evaluation Debugging is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Evaluation Debugging in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/adk-python/sub-skills/evaluation-debugging. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Evaluation Debugging?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Evaluation Debugging?

As many as you like. As long as a model supports skills, you can use Evaluation Debugging with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Evaluation Debugging AI skill free?

Yes. It is published on GitHub by VectorSpaceLab under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Eval Harness

affaan-m

CommunityPopular

Framework formal de evaluación para sesiones de Claude Code que implementa principios de desarrollo orientado a evals (EDD)

Tdd Workflow

Usar este skill al escribir nuevas funcionalidades, corregir bugs o refactorizar código. Aplica el desarrollo guiado por pruebas con 80%+ de cobertura incluyendo pruebas unitarias, de integración y…

Angular Developer

Angular コードを生成し、アーキテクチャガイダンスを提供します。プロジェクトの作成、コンポーネント、またはサービスを作成するとき、または反応性（シグナル、linkedSignal、リソース）、フォーム、依存性注入、ルーティング、SSR、アクセシビリティ（ARIA）、アニメーション、スタイリング（コンポーネントスタイル、Tailwind CSS）、テスト、または CLI ツール作成のベスト…

Anatomy

Explain an idea as an interactive isometric machine figure.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

Evaluation Debugging

Route Common Requests

Boundaries

Safe Defaults

Quick Decision Guide

Bundled Reference Map

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now