PolymathWizard

BHIL Colophon Spec — Testing skill for Claude Code

Testing community

Frameworks die before productization because abstraction is manual and unscheduled.

How to install BHIL Colophon Spec

This entry records only its repository, not the path inside it, so there is no exact command to give. Open PolymathWizard/BHIL-Colophon-Spec and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What BHIL Colophon Spec does

Frameworks die before productization because abstraction is manual and unscheduled. The colophon makes it a byproduct of delivery: every skill run deposits KPIs, tasks, and promotion candidates as end matter, routed under a capability grant. Spec + reference implementation + 47 conformance tests.

Alternatives in Testing

Test Runner — Run Python tests with pytest, unittest, or other testing frameworks 23.4k ★
Notion Spec To Implementation — Convert Notion specs into linked implementation plans and tasks 14.6k ★
Debug — You are tasked with helping debug issues during manual testing or implementation 10k ★

README

BHIL Colophon Specification

**An append-only trailing metadata block for `SKILL.md` files and the artifacts they produce.** *Front matter is a promise. End matter is a receipt.*

[![Quality gate](https://github.com/PolymathWizard/BHIL-Colophon-Spec/actions/workflows/quality-gate.yml/badge.svg)](https://github.com/PolymathWizard/BHIL-Colophon-Spec/actions/workflows/quality-gate.yml) [![Spec version](https://img.shields.io/badge/spec-v0.2%20draft-blue)](CHANGELOG.md) [![Code MIT](https://img.shields.io/badge/code-MIT-green)](LICENSE) [![Content CC BY 4.0](https://img.shields.io/badge/content-CC%20BY%204.0-green)](LICENSE-CONTENT)

What this is

Frameworks and ideas die before productization. The abstraction step — turning a one-off client engagement into a named, reusable, sellable asset — is manual, unscheduled, and loses every fight against billable work.

Front matter cannot fix this, because front matter is written *before* the engagement runs. At that moment you do not yet know what pattern you found, what it cost, or whether it is worth promoting.

The colophon is written *after*. Every skill run deposits, at the bottom of its own output, the structured residue of what it just learned: KPIs, follow-up tasks, and promotion candidates routed straight into the Framework Master Table. **Productization stops being a separate task and becomes a byproduct of delivery.**

This repository contains the specification, a working reference implementation, JSON Schemas, conformance requirements, and validated examples.

Key ideas


Extraction on delivery	Every run deposits promotion candidates with evidence attached. Nobody has to remember. → Rationale
The allocation rule	Front matter carries only what decides whether to trigger. It is charged on every conversation; end matter is free until read. → Tokenomics
The capability split	Front matter declares the rout