DevTools Directory

Inspect, validate, and debug the data structure behind a directory site. Paste a JSON entry and the tool checks for required fields, schema mismatches, broken URLs, and duplicate slugs against your full dataset.

The validator runs on a configurable schema — set which fields are required, which are optional, and which value patterns count as valid. Common rules: URLs must start with https://, slugs must be kebab-case, descriptions must be 50-300 characters. Run a batch validation across your whole directory to find every entry that breaks any rule.

For live debugging, the inspector can fetch any URL in your directory and show its current response code, redirect chain, and final URL. Useful for catching link rot before it shows up as 404s in front of real users.

The tool's check phases: schema validation (does the entry have all required fields with the right types), pattern validation (do strings match expected regexes — kebab-case slugs, https URLs, ISO country codes), uniqueness check (is the slug duplicate of any other entry, fuzzy or exact), URL liveness (does the entry's URL respond with a 200 today), and metadata fetching (when you provide a URL, the tool can pull the page's title, og:title, og:description, og:image to compare against the entry's own description). Each phase is independent — run only what you need.

Worked example: a 500-entry AI tools directory. Run batch validation. Schema reports: 3 entries missing required 'category' field, 12 entries with descriptions under 50 chars (too short for SEO), 1 entry with a non-https URL. Pattern check: 4 entries with slugs in camelCase instead of kebab-case (legacy data). Uniqueness check: 2 near-duplicates ('open-ai-codex' and 'openai-codex' point to the same product). URL liveness: 18 entries return non-200 responses — 7 are 301 redirects to new URLs (update the entry to the canonical URL), 11 are 404s (remove or replace). One pass of this surfaces a multi-hour cleanup task in 15 seconds and gives you a structured remediation list.

Where it gets tricky: rate-limiting on the URL liveness checks. Hitting 500 URLs in quick succession from a single IP gets you blocked by some hosts and skews results to false-negatives. The tool throttles to 1-2 req/sec by default and respects Retry-After headers. For directories under 100 entries this is fine; for larger sets, run validation overnight and accept the wait. Also: liveness checks are point-in-time. A site that's up today can 404 tomorrow. Schedule recurring validation (weekly) and capture the diff — entries that started failing this week are your active rot list. Manual remediation never scales; automated detection plus weekly review does.