Charcuterie: The Unicode Visual Similarity Explorer
Meta Description: Discover Charcuterie, the visual similarity Unicode explorer that helps developers and designers find lookalike characters. A complete guide to features, use cases, and tips.
TL;DR
Charcuterie is a browser-based Unicode explorer that lets you find visually similar characters across different Unicode blocks. Whether you're a developer hunting down homograph attacks, a designer looking for typographic alternatives, or a linguist exploring script relationships, this tool slices through the complexity of Unicode's 149,000+ characters to surface the ones that look alike. It's free, fast, and surprisingly deep — but it has a learning curve worth understanding before you dive in.
Key Takeaways
- Charcuterie is a specialized Unicode tool focused on visual similarity between characters, not just code point relationships
- It's particularly valuable for cybersecurity professionals identifying homograph/IDN spoofing attacks
- Designers and typographers use it to find Unicode lookalikes for creative or technical purposes
- The tool covers characters from Latin, Cyrillic, Greek, Arabic, CJK, and dozens of other scripts
- Visual similarity is algorithmically computed — results aren't perfect, but they're genuinely useful
- Free to use with no sign-up required; open-source components make it extensible
What Is Charcuterie? A Unicode Explorer Built Around Looks
If you've ever squinted at a URL and wondered whether that "a" is actually an "а" (spoiler: the second one is Cyrillic), you've already encountered the problem that Charcuterie – the visual similarity Unicode explorer was built to solve.
Unicode is the universal character encoding standard that underpins virtually every modern computing system. With over 149,000 characters spanning 161 scripts as of Unicode 15.1, it's an enormous, sprawling system. Most tools that explore Unicode organize characters by code point, block, or category — logical groupings that make sense to engineers but tell you nothing about how characters look.
Charcuterie flips that paradigm. Instead of asking "what script is this character from?", it asks "what other characters look like this one?" The name is a playful nod to the art of slicing and arranging — in this case, slicing through Unicode's complexity to surface characters that share visual DNA.
[INTERNAL_LINK: Unicode basics and character encoding guide]
Why Visual Unicode Similarity Actually Matters
Before diving into how Charcuterie works, it's worth understanding why this type of tool exists in the first place. The use cases are more varied — and more critical — than you might expect.
1. Cybersecurity: Homograph Attacks and IDN Spoofing
This is arguably the most high-stakes use case. Internationalized Domain Names (IDNs) allow non-Latin characters in URLs, which opened the door to homograph attacks — where a malicious actor registers a domain using characters that look identical to a legitimate domain but are technically different.
The classic example: аррӏе.com vs apple.com. The first uses Cyrillic characters. Your eye almost certainly can't tell the difference. A phishing campaign built around this could deceive even security-savvy users.
Security researchers and penetration testers use visual similarity explorers to:
- Enumerate possible homograph variants of a target domain
- Test whether security tools correctly flag lookalike domains
- Build blocklists of visually confusable character pairs
[INTERNAL_LINK: IDN homograph attacks explained]
2. Typography and Design
Designers sometimes need a specific shape that doesn't exist in the character set they're working with, or they want to understand why certain fonts render certain characters similarly. Charcuterie helps typographers:
- Find Unicode characters that approximate a desired glyph shape
- Understand cross-script visual relationships
- Identify characters that may cause rendering ambiguity in multilingual layouts
3. Linguistics and Script Research
Scholars studying script evolution often find that characters across different writing systems share visual roots or coincidental similarities. A tool that surfaces these relationships visually — rather than etymologically — offers a different lens on script history and development.
4. Software Internationalization (i18n) Testing
When internationalizing software, developers need to test how their UI handles characters from many different scripts. Finding characters that stress-test rendering engines — particularly ones that look similar but have different directionality, combining behavior, or glyph complexity — is a legitimate QA use case.
How Charcuterie Works: Under the Hood
Understanding the mechanics of Charcuterie helps you use it more effectively and interpret its results with appropriate nuance.
Visual Similarity Algorithms
Charcuterie doesn't use human-curated similarity lists (though some Unicode standards, like the Confusables data from Unicode Technical Report #39, inform the field). Instead, it computes similarity algorithmically, typically by:
- Rendering characters as bitmaps at a standardized size and font
- Comparing pixel distributions using image similarity metrics (often variants of structural similarity or feature hashing)
- Scoring pairs and surfacing the highest-scoring matches
This approach has real strengths: it catches similarities that human curators might miss, and it's scalable across the entire Unicode range. But it also has limitations — results depend heavily on the reference font used, and some algorithmically "similar" characters may look quite different in practice depending on the typeface.
The Interface: What You're Actually Looking At
The Charcuterie interface is clean and deliberately minimal. You:
- Input a character (by typing, pasting, or entering a code point)
- Set a similarity threshold (how strict the matching should be)
- Browse results organized by visual similarity score
Results show the matched character, its Unicode code point, its official Unicode name, its script block, and its similarity score. You can click into any result to use it as a new search seed — which is where the tool becomes genuinely exploratory and almost rabbit-hole-inducing.
Practical Walkthrough: Using Charcuterie Effectively
Let's walk through a real use case to make this concrete.
Scenario: Security Audit of a Brand Domain
Say you're responsible for protecting the domain secure-login.com for your company. You want to know what homograph variants an attacker might register.
Step 1: Enter the letter e into Charcuterie.
Step 2: Review visually similar characters. You'll likely find:
-
е(U+0435, Cyrillic Small Letter Ie) -
ė(U+0117, Latin Small Letter E with Dot Above) -
ẹ(U+1EB9, Latin Small Letter E with Dot Below) -
ℯ(U+212F, Script Small E) - Several more from mathematical and letterlike Unicode blocks
Step 3: Repeat for each character in your domain. Build a matrix of variants.
Step 4: Cross-reference with domain registrar lookups to see which variants are already registered (potentially by squatters or bad actors).
This workflow, which used to require manual research through Unicode charts, takes minutes with the right tool.
Pro Tips for Getting the Most Out of Charcuterie
- Adjust the similarity threshold carefully. Too strict and you'll miss meaningful matches; too loose and you'll be buried in noise. Start at 80% similarity and adjust from there.
- Consider font dependency. If your application uses a specific font, characters that look identical in Charcuterie's reference font may look distinct in yours. Always verify in context.
- Use it alongside Unicode TR#39 Confusables. The official Unicode confusables dataset is more conservative but carries authoritative weight. Charcuterie catches things TR#39 misses, and vice versa.
- Export results for downstream use. If you're building a blocklist or doing systematic research, export character lists rather than manually copying results.
Charcuterie vs. Other Unicode Tools: How Does It Compare?
There are several Unicode exploration tools available. Here's an honest comparison:
| Tool | Visual Similarity | Code Point Search | Script Filtering | Free | Open Source |
|---|---|---|---|---|---|
| Charcuterie | ✅ Core feature | ✅ | ✅ | ✅ | ✅ |
| Unicode Character Table | ❌ | ✅ | ✅ | ✅ | ❌ |
| Compart Unicode | ❌ | ✅ | ✅ | ✅ | ❌ |
| Unicode Confusables (TR#39) | ⚠️ Limited | ✅ | ⚠️ Limited | ✅ | ✅ |
| Shapecatcher | ✅ Draw-to-find | ❌ | ❌ | ✅ | ❌ |
Honest assessment: No single tool does everything. Charcuterie excels specifically at systematic visual similarity exploration. Shapecatcher is better if you're trying to identify an unknown character by drawing it. The Unicode Consortium's own confusables data is more authoritative for security-critical applications but far less comprehensive.
For most developers and researchers, the ideal workflow combines Charcuterie with Unicode Inspector for detailed character metadata.
[INTERNAL_LINK: Best Unicode tools for developers in 2026]
Limitations and Honest Caveats
No tool review is complete without an honest look at shortcomings. Charcuterie has several worth knowing:
Font Dependency Is Real
The similarity scores are computed against a specific reference rendering. Characters that score 95% similar in Charcuterie may look noticeably different in your application's chosen typeface. This is particularly true for characters from less-common scripts where font support is inconsistent.
Coverage Gaps in Complex Scripts
Characters from scripts with complex shaping rules — Arabic, Devanagari, Tibetan — are harder to compare visually in isolation because their appearance changes dramatically based on context (joining behavior, conjunct forms, etc.). Charcuterie's results in these script areas should be treated as directional, not definitive.
Not a Security Tool by Itself
While Charcuterie is valuable for security research, it shouldn't be your only defense against homograph attacks. Proper IDN handling at the browser/DNS level, certificate transparency monitoring, and domain monitoring services are all part of a complete strategy.
Performance at Scale
If you need to process thousands of characters programmatically, the browser interface isn't the right tool. Look for Unicode confusables libraries in your language of choice for bulk processing.
Who Should Use Charcuterie?
Definitely use it if you are:
- A security researcher or penetration tester working on domain/phishing analysis
- A developer building systems that need to handle or detect visually similar Unicode input
- A typographer or font designer exploring cross-script character relationships
- A linguist or Unicode enthusiast who enjoys exploring script systems
You might find it less useful if you are:
- Looking for a general-purpose Unicode reference (use Compart or the Unicode Character Database directly)
- Needing programmatic bulk processing (use a library instead)
- Working primarily with a single script where visual similarity is less ambiguous
The Broader Context: Unicode Visual Similarity in 2026
As of April 2026, Unicode 16.0 has added further characters to an already vast standard. The proliferation of emoji, the inclusion of more historical scripts, and the ongoing expansion of mathematical and technical symbols mean the visual similarity problem is getting more complex, not less.
At the same time, AI-assisted font rendering and increasingly sophisticated phishing detection have changed the landscape. Browser vendors have improved IDN display policies, and major registrars have tightened rules around mixed-script domain registration. But the fundamental challenge — that humans are terrible at distinguishing visually similar characters at a glance — hasn't changed.
Tools like Charcuterie remain essential precisely because the human visual system is the vulnerability. Technology can patch code, but it can't rewire how our eyes process letterforms.
[INTERNAL_LINK: Unicode security best practices for web developers]
Getting Started: Your First Steps with Charcuterie
- Visit the tool in your browser — no installation or sign-up required
- Start with a character you know well — your own name is a good seed
- Explore the results at 85% similarity to get a feel for what "similar" means in practice
- Try a security-focused search — enter a character from your company's domain and see what comes up
- Bookmark the tool for whenever you encounter a suspicious-looking character in a URL, email, or document
Conclusion and CTA
The Charcuterie visual similarity Unicode explorer fills a genuine gap in the Unicode tooling ecosystem. It's not trying to be everything — it's laser-focused on one problem (visual similarity) and solves it well. Whether you're hardening your organization's security posture against homograph attacks, doing serious typographic research, or just satisfying a healthy curiosity about how the world's writing systems relate to each other visually, it's a tool worth having in your toolkit.
Ready to explore? Open Charcuterie in your browser right now and search for the letter in your name that you think is most unique. You might be surprised how many Unicode characters are waiting to impersonate it.
If you found this guide useful, consider sharing it with your security team or developer community — the more people understand visual Unicode similarity, the harder it becomes for bad actors to exploit it.
[INTERNAL_LINK: Subscribe to our newsletter for more developer tools deep-dives]
Frequently Asked Questions
What exactly is a "visual similarity Unicode explorer"?
A visual similarity Unicode explorer is a tool that finds Unicode characters that look alike, regardless of their underlying code points, script blocks, or semantic meaning. Unlike standard Unicode databases that organize characters by encoding or language family, these tools use visual/image-based comparison to surface characters that a human eye might confuse. Charcuterie is one of the most capable tools in this category.
Is Charcuterie safe to use for security-critical work?
Charcuterie is a useful research and discovery tool for security work, particularly for identifying potential homograph attack vectors. However, it shouldn't be your sole defense. For production security systems, combine it with the official Unicode Confusables dataset (TR#39), proper IDN handling in your DNS/browser stack, and dedicated domain monitoring services. Think of Charcuterie as a research accelerator, not a complete security solution.
How is visual similarity calculated in Charcuterie?
Charcuterie renders characters using a reference font and computes similarity based on the visual appearance of the resulting glyphs — essentially comparing how the pixels are distributed. The exact algorithm involves bitmap comparison techniques similar to image hashing or structural similarity indices. Because results are font-dependent, characters that score as highly similar may look more distinct in different typefaces, especially for less common scripts.
Can I use Charcuterie programmatically or via an API?
The primary interface is browser-based, which limits programmatic use. For bulk processing or integration into automated workflows, you'll be better served by Unicode confusables libraries available in most major programming languages (Python's confusable_homoglyphs package is a popular option). Charcuterie's open-source components may also be adaptable for custom implementations — check the project repository for licensing and reuse options.
What's the difference between Charcuterie and the Unicode Confusables dataset?
The Unicode Confusables dataset (from Unicode Technical Report #39) is an officially maintained, human-curated list of character pairs that are visually similar. It's authoritative and conservative — every entry has been reviewed. Charcuterie's algorithmically generated similarity scores are broader and catch more potential matches, including ones not in TR#39, but they're also less vetted. For security applications, TR#39 is the gold standard; Charcuterie is better for exploratory research where you want comprehensive coverage over conservative precision.
Top comments (0)