error-handling-audit

Audit TypeScript error-handling discipline across throw hygiene, async handling, React boundaries, and observability.

3|5|Updated Apr 26, 2026
One-click install
npx skills add https://github.com/BenSheridanEdwards/ArchitectPlaybook --skill error-handling-audit-bensheridanedwards
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: error-handling-audit
Source: https://github.com/BenSheridanEdwards/ArchitectPlaybook/tree/main/error-handling-audit
Command: npx skills add https://github.com/BenSheridanEdwards/ArchitectPlaybook --skill error-handling-audit-bensheridanedwards

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Codebases often swallow errors silently, throw untyped values, miss error boundaries, or lack production observability, and these structural gaps are hard to spot in manual review. This Skill statically audits a TypeScript codebase against an opinionated four-layer error-handling baseline and reports concrete violations with file and line references. ## Core Features & Use Cases - Four-layer audit: Covers throw and catch hygiene, async and network error handling, React error boundaries, and logging and observability, preceded by a diagnostic snapshot of try/catch counts, async style, and detected tooling. - Structured findings contract: Writes findings.md, findings.json (schema 2.0.0), snapshot.md, and metadata.json to .architect-audits/error-handling-audit/ for scoring and roll-up into a Repository Quality Score. - Optional implementation plan: After reporting, asks whether to generate a descriptive Markdown plan detailing which catch blocks, call sites, boundaries, and observability primitives to fix. - Use Case: Run /error-handling-audit on a Next.js app before a release to surface empty catch blocks, fetch calls without timeouts, missing route-level error boundaries, and an unconfigured Sentry sample rate, then generate a remediation plan. ## Quick Start Run /error-handling-audit in the root of a TypeScript project to produce a findings report and optionally generate an implementation plan for the gaps.

Frequently Asked Questions about error-handling-audit

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

FAQPage Schema
How do I audit error handling in a TypeScript codebase?

Run /error-handling-audit from the project root. It statically checks throw and catch hygiene, async and network error paths, React error boundaries, and logging setup, then writes findings to .architect-audits/error-handling-audit/ and offers an implementation plan.

What does the error-handling audit check in React applications?

It checks for a root error boundary, route- or feature-level boundaries, fallback UI with recovery actions, forwarding errors to a reporting service, Suspense paired with boundaries, no try/catch around JSX renders, and mutation errors handled at the call site.

Does the error-handling audit work without React?

Yes. Layers 1, 2, and 4 apply to any TypeScript project. When React is not detected, every Layer 3 check is emitted as not-applicable with a null status so it does not distort the summary or quality score.

Can the audit query Sentry or production error data?

No. The audit is static-only by design and never queries error-reporting services or executes code. Runtime error frequency belongs to observability tooling like Sentry or Datadog, not to this architectural audit.

Why does the audit report not-applicable for my JavaScript project?

The skill supports TypeScript projects only. If tsconfig.json is missing, it writes a canonical not-applicable audit where every catalog check appears once with a null status rather than guessing results.

Does the error-handling audit modify my source code?

No. It only writes reports inside .architect-audits/error-handling-audit/. The optional implementation plan is descriptive Markdown listing fixes; it never edits source files, installs packages, or commits to git.