claude

test-harness

Retrofit a high-integrity test suite onto an untested codebase: seam-aware, risk-ranked, oracle-grounded, and mutation-gated. Enforces Generator/Verifier role separation and the net-immutability and oracle-independence invariants mechanically.

0 on GitHubby Bradley Smockv0.1.0MITupdated 1 month agodevelopment
/

Plugin test-harness from marketplace bradleysmock-plugins in bradleysmock/claude-harness. Commands read from the repo's manifest — 1 month ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Slash commands (12)

  • /test-harness:characterize

    Phase 3 — generate characterization tests that record current behavior as a refactoring net, verify green against unmodified source, then freeze the net (read-only).

  • /test-harness:contract-tests

    Phase 5 — lock module/service seams with contract tests covering input acceptance, output shape, error contract, and invariants.

  • /test-harness:elicit

    Oracle Stage 5 — turn weak/conflicted high-risk units into a batched human review packet (confirm/deny/correct), then record answers as authoritative claims.

  • /test-harness:harvest

    Oracle Stage 0 — harvest candidate oracle sources for a unit (git/bug history, schemas, call sites, in-code contracts, standards). Deterministic gather; no judgement.

  • /test-harness:infra

    Phase 1 — stand up the test harness end-to-end with one smoke test and the mutation engine wired; prove the loop closes before scaling.

  • /test-harness:mutation-gate

    Phase 6 — the binding quality gate. Run diff-scoped mutation testing; loop verifier-triage → generator-kills until kill-rate target or all survivors are provably equivalent.

  • /test-harness:oracle-extract

    Oracle Stages 1-3 — extract normalized claims from harvested candidates, run the verifier independence gate, corroborate, and write the oracle ledger.

  • /test-harness:quality-audit

    Run the test-quality assessment (Q1-Q6) and produce a two-axis scorecard that never averages fault detection against oracle validity.

  • /test-harness:risk-rank

    Phase 2 — rank modules into risk waves using git churn, blast radius, complexity, and defect density. Never deprioritize high-risk-but-hard-to-test code.

  • /test-harness:run-module

    Drive one module end-to-end through the full pipeline (P0 → P6), pausing at the two human gates. Use this for the walking-skeleton pilot.

  • /test-harness:seam-map

    Phase 0 — build a seam map for a module: classify dependencies, seam types, and refactor cost; gate with the verifier.

  • /test-harness:unit-tests

    Phase 4 — refactor for testability under the frozen net, then write unit tests whose assertions cite an independent oracle. The only phase where assertions encode correctness.

Subagents (2)

  • generator

    Test generator (role G). Invoke to produce analysis (seam maps, risk rankings, oracle claims) and test code (characterization nets, unit tests, contract tests, mutation kills). Bound by oracle-sourcing discipline; its output is always gated by the verifier.

  • verifier

    Adversarial test verifier (role V). Invoke to gate any artifact the generator produced — seam maps, characterization nets, unit/contract tests, mutation survivors, oracle-claim independence. Never writes or edits code; only critiques and gates.

Hooks (1)

  • PreToolUse

    ${CLAUDE_PLUGIN_ROOT}/scripts/net-guard.sh

View source on GitHub

More from bradleysmock/claude-harness

All 5 plugins →

Unified Claude Code harness: polyglot gate pipeline (lint/type-check/tests/security with diff-based repair loop and BM25 failure memory) + two-checkpoint SDLC workflow (problem → design critic → spec → build → post-build critic → deliver). Twenty-nine expert review panels covering languages, frontend frameworks, security (identity + cryptography), data/ML engineering, and operations. No API key required.

11 skills · 31 commands · 3 agents · 1 MCP server · hooks

0

Structured two-checkpoint SDLC harness: slash commands, polyglot validation hooks, expert review panels, and a working agreement.

12 commands · 1 agent · hooks

0

harness

claude

MCP server wrapping the harness-full LLM coding harness: spec generation, polyglot gates, failure memory, and multi-spec task orchestration.

8 commands · 1 MCP server

0

Harness without an API key: Claude Code is the orchestrator, Python provides gates, memory, and DAG tools via MCP. Requires only a Claude Code subscription.

6 commands · 1 MCP server

0

Similar plugins

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.

5 commands

HKUDS/CLI-Anything 58,665

Automates browser interactions for web testing, form filling, screenshots, and data extraction

12 skills

vercel-labs/agent-browser 50,373

Use Alibaba Open Code Review from local Codex.

2 skills

alibaba/open-code-review 50,068

slidev

claude

Claude Code skill for creating and presenting Slidev slidedecks

1 skill

slidevjs/slidev 49,452

Maintain test-harness?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins5 plugins