Agent Evaluation

Not yet reviewedagentsPrompt / Rule# agents # skill # ai

Framework for testing and evaluating AI agent performance.

Not yet reviewed (manual audit pending)

This skill is cataloged from open community sources and has not yet completed full manual verification. Inspect repository code prior to local execution.

Install Agent Evaluation in Google Antigravity

Antigravity loads skills from .agents/skills/ in your project or ~/.gemini/config/skills/ for all projects (official docs). Full guide

EasiestOption 1: paste this to your Antigravity agent

Install the "Agent Evaluation" skill for Google Antigravity.
Follow the steps at https://www.antigravityskills.org/install/agent-evaluation.md (source: https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/agent-evaluation).
Before running anything, show me what you will download and whether it includes scripts, and wait for my OK.
Install it into this project (.agents/skills/) unless I say global. When done, check that Antigravity lists it as an installed skill.

The agent reads the install file, shows you what it will download and waits for your OK.

Option 2: install it yourself

1. Download the skill folder
```
git clone --depth 1 https://github.com/sickn33/antigravity-awesome-skills.git /tmp/ag-skills
```
The skill lives in skills/agent-evaluation/ of the source repository.
2. Copy it into this project (workspace skill)
```
mkdir -p .agents/skills && cp -r /tmp/ag-skills/skills/agent-evaluation .agents/skills/
```
Run from your project root. Antigravity loads workspace skills from .agents/skills/.
3. Or install it for all projects (global skill)
```
mkdir -p ~/.gemini/config/skills && cp -r /tmp/ag-skills/skills/agent-evaluation ~/.gemini/config/skills/
```
Global skills live in ~/.gemini/config/skills/. For the Antigravity CLI use ~/.gemini/antigravity-cli/skills/ instead.
4. Check that Antigravity sees it
Antigravity 2.0 app: start a new conversation and ask "What skills do you have?" (the app has no /skills command; typing / and the skill name also finds it). Antigravity CLI: run /skills; in a session that was already open, run /skills reload first. The agent loads the skill when your request matches its description.

Using Claude Code, Cursor or Codex? See other assistants below.

What is Agent Evaluation?

Evaluate agent behavior with versioned cases and explicit verifiers. Use when comparing agent or prompt changes, reproducing failures, or running agent regression tests.
Description from the upstream SKILL.md

What the Agent Evaluation SKILL.md covers

✓Evaluation procedure
✓Example: changed tool argument handling
✓Worked uncertainty example
✓Optional architecture patterns

Source and files

Repository: sickn33/antigravity-awesome-skills (MIT)
Upstream risk label: safe
Added upstream: 2026-02-27
SKILL.md size: 7.6 KB
Files in the skill folder: SKILL.md · references
Listing synced: 2026-10-05 from commit bdacd76

Installation & Setup Guide

Agent Rules / Prompt Setup

RecommendedQuick install command

git clone --depth 1 https://github.com/sickn33/antigravity-awesome-skills.git /tmp/ag-skills && mkdir -p ~/.gemini/config/skills && cp -r /tmp/ag-skills/skills/agent-evaluation ~/.gemini/config/skills/ && rm -rf /tmp/ag-skills

▶Step-by-step install

Step 1: Create skills directory

mkdir -p ~/.gemini/config/skills

Step 2: Clone the skills repository

git clone --depth 1 https://github.com/sickn33/antigravity-awesome-skills.git /tmp/ag-skills

Step 3: Copy this skill

cp -r /tmp/ag-skills/skills/agent-evaluation ~/.gemini/config/skills/

Step 4: Clean up temporary files

rm -rf /tmp/ag-skills

Security Review Status

Not yet reviewed (manual audit pending)

This skill is cataloged from open community sources and has not yet undergone full manual line-by-line security verification by our team.

Recommended Verification Steps Prior to Execution:

Inspect the source repository files before adding to your local AI tool config.
Avoid passing production API credentials or sensitive filesystem permissions unless explicitly required.
Run initial commands inside a sandboxed terminal or container environment.

Source repository link for manual inspection: https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/agent-evaluation

Operational Notes & Known Limitations

Known Limitations

•Illustrative 80/90% thresholds and score weights in the architecture sketches are not universal merge/deploy rules; define project-specific criteria and keep critical failures separate.
•A small-sample chi-squared comparison or absence of significance does not prove equivalence; use a method suited to counts, pairing and multiple comparisons.
•Exceptions are not automatic safe rejections, and test retries must not erase the first failure.
•Similarity to a retrieved answer may be legitimate RAG behavior; leakage depends on what the evaluation permits the agent to know.
•LLM judges do not substitute for real user feedback, and output truncation does not remove private data. Use synthetic or authorized redacted inputs with bounded retention.