adk-eval-guide

adk-eval-guide is a skill for Claude Code from kumaran-is/claude-code-onboarding. It costs 110 tokens per session (1,616 once invoked), scanned A, original, MIT.

A guide for evaluating Google ADK agents with evaluation sets, which are saved collections of test conversations and expected behavior. It documents eight evaluation criteria and explains how to configure and investigate results.

In plain words
What is it for?
Use it to create evaluation cases, choose metrics, run the ADK evaluation command, test tool calls and conversations, and diagnose failed results.
Why use it?
It avoids unreliable and costly tests against the live Gemini API and makes agent behavior checks repeatable in development and CI.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Good fit Use it to create evaluation cases, choose metrics, run the ADK evaluation command, test tool calls and conversations, and diagnose failed results.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/kumaran-is/claude-code-onboarding/adk-eval-guide
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add kumaran-is/claude-code-onboarding --skill adk-eval-guide
Clone the repo
git clone --depth 1 https://github.com/kumaran-is/claude-code-onboarding

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for adk-eval-guide

README.md
[![agentmods](https://agentmods.dev/badge/skills/kumaran-is/claude-code-onboarding/adk-eval-guide/github.svg)](https://agentmods.dev/skills/kumaran-is/claude-code-onboarding/adk-eval-guide)
Your own site
<a href="https://agentmods.dev/skills/kumaran-is/claude-code-onboarding/adk-eval-guide"><img src="https://agentmods.dev/badge/skills/kumaran-is/claude-code-onboarding/adk-eval-guide/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for adk-eval-guide

Your own site · 80×15
<a href="https://agentmods.dev/skills/kumaran-is/claude-code-onboarding/adk-eval-guide"><img src="https://agentmods.dev/badge/skills/kumaran-is/claude-code-onboarding/adk-eval-guide.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 110 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,616 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00110 $0.01616
Opus 5 $0.00055 $0.00808
Sonnet 5 $0.00022 $0.00323
Haiku 4.5 $0.00011 $0.00162

Measured 6d ago against content hash 434ded9ff1d4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

adk-eval-guide scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/adk-eval-guide/SKILL.md · 122 lines

How it starts

The opening of the file, as written. The whole thing — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ADK Evaluation Guide

Iron Law

NEVER evaluate agents against the live Gemini API in unit tests. Use adk eval with evalsets. Hitting the real API in eval makes results non-deterministic, incurs cost, and breaks CI. Always run evaluations through the ADK eval CLI with a defined evalset and config file.

Reference Files

File Contents
reference/criteria-guide.md Complete metrics reference — all 8 criteria, match types, custom metrics, judge model config
reference/user-simulation.md Dynamic conversation testing — ConversationScenario, user simulator config, compatible criteria
reference/builtin-tools-eval.md google_search and model-internal tools — trajectory behavior, metric compatibility
reference/multimodal-eval.md Multimodal inputs — evalset schema, built-in metric limitations, custom evaluator pattern

Process

  1. Define criteria — choose metrics based on your goal (see Choosing the Right Criteria below); read reference/criteria-guide.md for all 8 options
  2. Write evalset — create evalset.json with eval cases; include intermediate_data.tool_uses for every turn that expects tool calls
  3. Write eval config — create eval_config.json with criteria thresholds; set match_type to IN_ORDER unless you need strict EXACT regression tests
  4. Run evaladk eval ./app <path_to_evalset.json> --config_file_path=<path_to_config.json> --print_detailed_results
  5. Interpret results — identify which criteria failed; match failure symptom to cause (see Common Eval Failure Causes below)
  6. Fix the agent — adjust prompts, tool descriptions, or agent instructions; do NOT widen thresholds to hide failures
  7. Fix the evalset if needed — if expected trajectory doesn't match real agent behavior due to model variance, switch match type or use rubric-based criteria
  8. Rerun eval — verify the fix resolves the specific failure without regressing other cases
  9. Expand coverage — once current cases pass, add more eval cases incrementally
  10. Repeat — expect 5–10 iterations per feature; each iteration makes the agent more reliable

Read the full file on GitHub · 122 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 122 lines · 110 tokens per session scan A 434ded9ff1d4

Subscribe to this mod's changes

adk-eval-guide is a skill published in the GitHub repository kumaran-is/claude-code-onboarding (35 stars, last pushed 2mo ago), licensed MIT. It adds 110 tokens to every session and 1,616 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Related

Other skills, from other repositories

cli-eval

Create and run evaluation suites, watch live benchmark progress, view scorecards, compare model performance, and integrate eval runs with CI workflows from the CLI.

diegosouzapw/OmniRoute · 34 tokens

model-merging

Merge multiple fine-tuned models using mergekit to combine capabilities without retraining. Use when creating specialized models by blending domain-specific expertise (math + coding + chat), improving performance beyond single models, or experimenting rapidly with model variants. Covers SLERP, TIES-Merging, DARE, Task…

davila7/claude-code-templates · 73 tokens

darwinian-evolver

Evolve prompts/regex/SQL/code with Imbue's evolution loop.

NousResearch/hermes-agent · 22 tokens

validate

Validate Semantica pipelines, extraction quality, graph schemas, and ontology consistency. Returns structured error/warning checklists. Uses PipelineValidator, PipelineBuilder.validatepipeline(), GraphValidator, and OntologyValidator. Sub-commands: pipeline, step, dependencies, extraction, graph, ontology, performance.

semantica-agi/semantica · 0 tokens

launching-evals

Run, monitor, analyze, and debug LLM evaluations via nemo-evaluator-launcher. Covers running evaluations, checking status and live progress, debugging failed runs, exporting artifacts and logs, and analyzing results. ALWAYS triggers on mentions of running evaluations, checking progress, debugging failed evals…

NVIDIA/Model-Optimizer · 115 tokens

nemo-automodel-recipe-development

Create and modify NeMo AutoModel training and evaluation recipes, including YAML structure, builders, and execution flow.

NVIDIA/skills · 31 tokens