skills.rest
dmmulroy
dmmulroyCOMMUNITY
Explore all skills by this author →
Website𝕏@dillon_mulroyGitHub

audit-better-result-dependents

Audits better-result API changes against Prisma and Better T Stack npm dependents for breakage.

1.9k|43|Updated Jan 9, 2026
Search Tags:#prisma#dependency audit#NPM#typeScript#api compatibility#Pull request#Breaking Changes
One-click install
npx skills add https://github.com/dmmulroy/better-result --skill audit-better-result-dependents
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: audit-better-result-dependents
Source: https://github.com/dmmulroy/better-result/tree/main/.agents/skills/audit-better-result-dependents
Command: npx skills add https://github.com/dmmulroy/better-result --skill audit-better-result-dependents

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

When changing the better-result library's public API, types, or runtime behavior, maintainers risk silently breaking downstream packages like Prisma SDKs and Better T Stack templates. This Skill automates checking whether a proposed change or PR breaks those known dependents before merging.

Core Features & Use Cases

  • Automated dependent auditing: Runs a script that resolves the latest published versions of @prisma/compute-sdk, @prisma/streams-server, @prisma/streams-local, create-better-t-stack, and @better-t-stack/template-generator from the npm registry, then greps their source for usage patterns affected by your change.
  • Flexible source resolution: Checks out the published gitHead from GitHub when available, and falls back to npm pack for private or unavailable repositories.
  • Structured breakage assessment: Guides distinguishing runtime-breaking, type-breaking, and safe changes, with a reporting format suitable for PR bodies.
  • Use Case: Before merging a PR that changes TaggedError.is guard semantics, run the audit to confirm none of the Prisma or Better T Stack dependents rely on the old behavior.

Quick Start

Ask the assistant to audit whether your current better-result changes break Prisma and Better T Stack dependents, optionally naming the specific identifiers or patterns you modified.

Frequently Asked Questions about audit-better-result-dependents

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

FAQPage Schema
How do I check if a better-result change breaks downstream dependents?▼

Run the audit script with identifiers touched by your change, such as `node .agents/skills/audit-better-result-dependents/scripts/audit-dependents.mjs isTaggedError`. It greps the latest published source of known Prisma and Better T Stack dependents for matching usage patterns.

Which npm packages does the better-result dependent audit cover?▼

The audit always checks five dependents: `@prisma/compute-sdk`, `@prisma/streams-server`, `@prisma/streams-local`, `create-better-t-stack`, and `@better-t-stack/template-generator`. These are resolved to their latest versions via npm registry metadata.

What happens if a dependent's GitHub repository is private or unavailable?▼

The script falls back to `npm pack` and audits the published package source instead. For example, `@prisma/compute-sdk` may point to a private repo, so its npm tarball is downloaded and searched directly.

How do I distinguish runtime-breaking from type-breaking changes in dependents?▼

Runtime breakage occurs when a dependent passes values that no longer satisfy a changed guard or shape. Type breakage occurs when a dependent references narrowed types, overloads, or exported names that changed. Dependents only creating `TaggedError` instances are typically safe.

What tools are required to run the dependent audit script?▼

The script requires Node.js, git, npm, and ripgrep (`rg`). It uses the npm registry API to resolve versions, git to clone public repos at their published gitHead, and ripgrep to search dependent source code for risk patterns.

Related Skills

Explore More Software Engineering Skills→
jmolzjmolz

source-command-validate

Run automated codebase validation steps with detailed status reporting.

Community
Intermediate
the-agency-aithe-agency-ai
39

pr-captain-land

Merge agent branches into main with version bumps and CI verification.

Official
Advanced
alexmenschalexmensch

semver

Classify code changes as major, minor, or patch for semantic versioning.

Community
Intermediate
watminwatmin

complectens

Identify layered named test helpers with per-layer deftests.

Community
Advanced