Reality Verification

This is about making sure your fixes actually work by reproducing the failure before you code and verifying it's gone after. It detects when you're trying to fix something versus adding something new, then guides you through running the right command (tests, types, CI logs, whatever) and documenting both states in a progress file. The VF task it generates forces you to compare before and after output explicitly. It also flags mock-heavy tests where you're just verifying mock interactions instead of real behavior. Honestly, this is the kind of disciplined workflow that feels tedious until it saves you from merging a "fix" that doesn't actually fix anything.

GitHub stars: 340
last updated: May 19, 2026

View on GitHub

Install

$ npx -y skills add tzachbon/smart-ralph --skill reality-verification --agent claude-code

Installs into .claude/skills of the current project.

Files

Reality Verification

For fix goals: reproduce the failure BEFORE work, verify resolution AFTER.

Goal Detection

Classify user goals to determine if diagnosis is needed. See references/goal-detection-patterns.md for detailed patterns.

Quick reference:

Fix indicators: fix, repair, resolve, debug, patch, broken, failing, error, bug
Add indicators: add, create, build, implement, new
Conflict resolution: If both present, treat as Fix

Command Mapping

Goal Keywords	Reproduction Command
CI, pipeline	`gh run view --log-failed`
test, tests	project test command
type, typescript	`pnpm check-types` or `tsc --noEmit`
lint	`pnpm lint`
build	`pnpm build`
E2E, UI	Playwright MCP browser tools
API, endpoint	WebFetch tool