원클릭으로
skill-behavioral-evals
Engineering guide for creating and running behavioral evals to verify agent decision logic and steerage.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Engineering guide for creating and running behavioral evals to verify agent decision logic and steerage.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | skill-behavioral-evals |
| description | Engineering guide for creating and running behavioral evals to verify agent decision logic and steerage. |
Behavioral evaluations (evals) are tests that validate the agent's decision-making (e.g., tool choice) rather than pure functionality. They are critical for verifying prompt changes, debugging steerability, and preventing regressions.
[!NOTE] Single Source of Truth: For core concepts, policies, running tests, and general best practices, always refer to evals/README.md.
appEvalTest (AppRig). See creating.md.evalTest (TestRig). See creating.md.USUALLY_PASSES.ALWAYS_PASSES (locks in regression).Seed the workspace with necessary files using the files object to simulate a realistic scenario (e.g., NodeJS project with package.json).
Audit agent decisions using rig.setBreakpoint() (AppRig only) or index verification on rig.readToolLogs().
Run single tests locally with Vitest. Confirm stability locally before relying on CI workflows.
Detailed procedural guides:
Use the following runbooks for deep-dive investigation and implementation.
| Capability | Reference File |
|---|---|
| Creating | creating.md |
| Fixing | fixing.md |
| Promoting | promoting.md |
| Running | running.md |
Professional bootstrapper for Vercel-style Agentic Repositories using compressed AGENTS.md manifests.
Persistent SQLite-based cognitive layer for cross-session interaction logging and insight distillation.
Design lead for C4 structural documentation, Mermaid.js diagrams, and Architecture Decision Records (ADRs).
Specialist in core AWS infrastructure: EC2 compute, IAM identity, VPC networking, and RDS/Aurora databases.
5-phase migration engine for transitioning workloads from Google Cloud Platform to AWS architecture.
AI/ML infrastructure expert for SageMaker fine-tuning, HyperPod cluster management, and dataset evaluation.