principle-prove-it-works

Verify completed work by directly inspecting real artifacts instead of proxies or self-reports.

6.6k|542|Updated Jan 23, 2026
One-click install
npx skills add https://github.com/cursor/plugins --skill principle-prove-it-works
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: principle-prove-it-works
Source: https://github.com/cursor/plugins/tree/main/pstack/skills/principle-prove-it-works
Command: npx skills add https://github.com/cursor/plugins --skill principle-prove-it-works

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Work declared done without direct verification has unknown correctness, and indirect signals like file mtimes, cached screenshots, or agent self-reports often mislead. This Skill enforces direct observation of the real artifact before any task is considered complete.

Core Features & Use Cases

  • Direct Verification Discipline: Check process liveness, actual values, and runtime behavior directly rather than through derived or cached state.
  • End-to-End Feature Testing: Build, run, and exercise the actual feature path, confirming data flows from input to output across integrations.
  • Delegation Auditing: Inspect the actual output artifact (git diff, file contents, runtime behavior) instead of trusting a delegate's summary.
  • Scripted Proof: Write deterministic re-runnable comparison scripts and keep their output as reviewable artifacts, committing them for large migrations.
  • Use Case: After an agent claims a refactor is complete, run the feature, inspect the git diff, and execute a script comparing old and new compiled output before approving.

Quick Start

After finishing the current task, verify it actually works by running the feature and inspecting the real output artifact rather than trusting the build or self-report.

Frequently Asked Questions about principle-prove-it-works

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify that a coding task is actually complete?

Verify completion by running the feature and exercising the actual code path, not just confirming it builds. Check that data flows from input to output, and for integrations test the full communication path end-to-end.

How to verify work delegated to an AI agent?

Verify delegated work by inspecting the actual output artifact such as the git diff, file contents, or runtime behavior rather than the agent's summary. Agents report what they intended, not always what happened.

Why is 'it compiles' not enough to declare a task done?

Building is necessary but not sufficient because compilation does not exercise runtime behavior. Indirect signals like file mtimes, output freshness, or cached screenshots can be stale or misleading compared to direct observation.

When should verification scripts be committed to the repository?

Commit verification artifacts only for large or complex work where the trail must be auditable later, such as a big port or migration. Most work only needs the artifact visible for the reviewer, not committed.

What should I check first when verification fails?

When verification fails, suspect the observation method before suspecting the system. Confirm you are reading the actual value directly rather than a cached or derived representation before concluding the system is broken.