Slug Generator — URL-Safe Slugs with Real Transliteration, in Bulk

Turn titles into clean URL slugs — with actual transliteration, not just stripping: diacritics (Héllo Wörld → hello-world), Cyrillic (Привет мир → privet-mir), Greek, symbol words (& → and, € → euro), and batch-unique suffixes when two titles collide. Up to 200 texts per run, online, by API, or as an agent tool via Apify MCP.

CMS imports, programmatic SEO pages, migrating a catalog, generating filenames — naive slugify functions silently eat non-ASCII titles into empty strings or duplicate URLs. This one is release-tested on exactly those cases.

What you get

• slug — URL-safe, separator-joined, trimmed, deduped within the batch (hello-world, hello-world-2)
• length — for URL-budget checks
• Options: separator (default -), lowercase (default true), maxLength (cut at a word boundary)
• Fail-soft: an input with no sluggable characters (emoji-only, punctuation-only) never fails the run — it returns {ok: false, error} and is never charged.

Input

{ "texts": ["10 Café Tips & Tricks", "Über uns", "Привет мир"], "separator": "-" }

Or a single string via text.

Output (real run)

[
  { "input": "Héllo Wörld — Ærøskøbing & Café №5!", "ok": true, "slug": "hello-world-aeroskobing-and-cafe-no5", "length": 36 },
  { "input": "Привет мир", "ok": true, "slug": "privet-mir", "length": 10 },
  { "input": "Hello World", "ok": true, "slug": "hello-world", "length": 11 },
  { "input": "Hello World", "ok": true, "slug": "hello-world-2", "length": 13 }
]

Pricing

$0.0004 per slug generated. No start fee. Unsluggable inputs are never charged. Slugging a 10,000-product catalog costs $4 across runs.

The only direct store incumbent measured (2026-08-07, nrithm/slug-generator) is free but single-input without transliteration per its listing; no paid comparable exists — pricing is cost-plus at the floor of our ladder.

Honest limits

• Transliteration covers Latin diacritics, Greek, Cyrillic, and common symbols. CJK, Arabic, Hebrew and other scripts are not transliterated (no meaningful ASCII mapping exists without a phonetic engine) — such input yields an uncharged {ok:false} record rather than a garbage slug.
• Uniqueness suffixes are per-batch: collisions across separate runs are yours to track (pass existing titles in the same batch to dedupe against them).
• maxLength trims at a separator boundary, so a very long first word can produce a slug shorter than the cap.

FAQ

How are duplicate titles handled? Within a run, the second identical slug gets -2, the third -3 — so a batch import never produces colliding URLs.

What happens to &, %, $? They become words: and, percent, dollar — like the slugs major CMSs produce, keeping meaning instead of dropping it.

Can I use underscores or no separator? Yes — separator accepts any string, including _; repeated separators are collapsed and trimmed.

Is the German ß handled? Yes — ß → ss, plus æ → ae, œ → oe, þ → th, and the full Latin diacritic set.

Why did some rows come back ok: false? Nothing sluggable survived (emoji/punctuation-only, or a script we don't transliterate). Recorded, never charged.

Use from code or AI agents

curl -s "https://api.apify.com/v2/acts/EliAI~slug-generator/run-sync-get-dataset-items?token=$APIFY_TOKEN" \
  -X POST -H 'Content-Type: application/json' \
  -d '{"texts": ["My First Blog Post!", "Über die Führung"]}'

Agents: connect Apify MCP and call the EliAI/slug-generator tool.

• Capability: URL-safe slug generation with transliteration + batch dedupe for one or many titles
• Required input: text (string) or texts (array); optional separator, lowercase, maxLength
• Returns: one record per input; slug is the deliverable
• Bounded: 200 texts per run; failures isolate per input
• Side effects: none