KNOLO / HUBPublishknolo.dev → Runtime
the3seus/prompt-eval

Prompt Eval Handbook

A queryable handbook for golden sets, LLM judges, pass@k, leakage, and prompt regression — written for people who ship prompts.

Query pack →Download
V5✓ Verified artifactPublisher-attestedApache-2.0v1.0.042 KB14 docsUpdated 1 day ago★ 0↓ 2
KNOWLEDGE CARD / README

Portable policy knowledge for local retrieval.

Prompt Eval Handbook is original evaluation guidance. It is not a vendor cookbook.

A prompt change without a frozen golden set is an anecdote, not a release.

The included agent metadata is never executed by Knolo Hub.

Intended use

Grounding prompt and model evaluation. Lookup when building a golden set, using a judge, or blocking a prompt change.

Out of scope

This pack does not call a model, score production traffic, or store API keys. It does not run a shell.

Sources

Sample questions

!
Contains agent metadata

This Knowledge Image contains prompts, tool policy, namespace grants, or other agent registry metadata.

Mounting the pack for retrieval does not execute tools. Applying these policies affects your application only if your host explicitly honors them.

Credentials must never be stored inside a pack.
Inspect registry →
NAMESPACE SUMMARY
01sampling2 docs→02judges2 docs→03regression2 docs→04metrics2 docs→05leakage2 docs→06goldens2 docs→07metadataView metadata→08agentsRegistry→09claimsGraph→
✓
Verified artifactContainer, digest, structure passed
◆
Publisher-attestedLicense, sources, rights supplied
◌
Not auditedIndependent quality signal

Artifact verification confirms integrity and format. It does not guarantee that the contained knowledge is factually correct.

Prompt Eval Handbook — Knolo Hub