agent-harness

SKILLFlusso di lavorocommunity
v0.0.0borgheiNOASSERTIONAggiornato 8 g faFonte →

Test and evaluation harness for AI agents — scenario suites, deterministic replay, regression diffing, cost and latency budgets. Use when agent quality is vibe-checked, before shipping a prompt or model change, or when evals drift.

Community-submitted skill. Not yet reviewed by the Forge team. Full prompt content may not be available.Request review →
481Stelle del repo
1Client
1Formati
8 g faUltimo aggiornamento
Skill
Autoreborghei
Versione0.0.0
LicenzaNOASSERTION
CategoriaFlusso di lavoro
Formatiskill.md
PromptNon pubblicato
Compatibilità
Claude✓ Supportato
Cursor—
Copilot—
ChatGPT—
Gemini—
Descrizione

Test and evaluation harness for AI agents — scenario suites, deterministic replay, regression diffing, cost and latency budgets. Use when agent quality is vibe-checked, before shipping a prompt or model change, or when evals drift.

Parole chiave
skillclaude