Relevio is a unique context‑aware, rule‑driven moderation engine that enables full and reliable autonomous story boards and forums.

Take away the burden of moderatating your website today.

How does it work?

Relevio is a new class of moderation engine employing Rule‑Driven Semantic Moderation, which uses the full power of large language models (LLM). This is very different from the established method of Category‑Based Classifier Moderation employed by systems such as Hive or Tisane.

Rule‑driven, not model‑driven

You define the rules in plain English. Relevio applies them consistently, using AI only as a semantic interpreter. There is no drift, no unpredictability and no black‑box behaviour.

Context‑aware by design

Every check can include context (the related story, prior comments, or reply chain). Relevio understands where a new post sits in the conversation, so it can enforce relevance and keep threads coherent.

Multimodal coherence

Relevio checks that images and text actually belong together, evaluating them as a single semantic unit. This prevents users from pairing misleading images with unrelated narratives and keeps your platform coherent, trustworthy, and harder to exploit.

Moderation before submission

Relevio is invoked from the browser or client. Content is checked before it ever reaches your server. This shifts the entire trust boundary to the client, dramatically reducing your attack surface and making it much easier to integrate.

Cryptographically verifiable decisions

Every approved submission is signed with an HMAC that binds the content, images, ruleset, and context together. Your backend accepts only trusted, tamper‑proof posts.

Minimal backend engineering

Your backend doesn’t need queues, pipelines, or human‑moderation interfaces. It simply verifies a fast, constant‑time HMAC and rejects anything that isn’t signed. All moderation latency happens in the client, not your infrastructure.

Instant feedback for users

Relevio gives writers clear, structured explanations and actionable suggestions before they submit. This reduces the irritation and distrust people feel when their content is rejected by a black‑box system. Instead of “your post was blocked,” users get transparent guidance that helps them fix issues quickly and stay engaged.

Built for autonomous story boards and forums

Relevio is the moderation engine built for self‑governing communities. Perfect for story walls, comment threads, and any platform where relevance and safety matter.

How does Relevio stack up?

Here's how Relevio compares with Category‑Based Classifier Moderation.

Dimension	Relevio	Category‑Based Classifier
Where moderation happens	Invoked from the browser — content is checked before submission	Server-side — content is ingested first, then classified
Security model	Pre-submission gatekeeping — your backend only receives approved, signed content	Post-submission filtering — harmful content still reaches your server
Backend engineering	Minimal — backend only verifies a fast HMAC; no queues, pipelines, or moderator UI	Heavy — requires queues, rate-limiting, pipelines, dashboards, and human oversight
Latency distribution	Latency sits in the client, keeping your backend fast and predictable	Latency sits in the server, requiring scaling and back-pressure handling
Context awareness	Yes — story, comment, and reply context included in every check	No — evaluates each message in isolation
Image + story consistency	Multimodal semantic checking — verifies that images match the story and flags contradictions	Independent classifiers — image and text are evaluated separately with no cross-consistency checks
Decision logic	Rule-driven — deterministic pass/fail based on your ruleset	Model-driven — probabilistic scores based on opaque model weights
Output format	Structured JSON with pass/fail, error description, and suggestion	Structured JSON with Category scores (e.g., toxicity %, spam %)
Explanation quality	Rule-specific, contextual explanations that tell users exactly which rule failed and how to fix it	Category-based explanations tied to fixed labels (e.g., bullying, hate); cannot reference custom rules or full context
Input size limit	Bounded by the LLM context window — can handle long stories and rich context	Max ~1024 characters for text moderation, limiting story length and context
Cryptographic trust	HMAC-signed decisions binding content, images, ruleset, and context	None — no verifiable proof of moderation integrity
Injection resistance	Strong — Malformed rulesets, and instruction attempts are rejected. Adversarial phrasing is readily detected by the LLM	Variable - Vulnerable to adversarial phrasing — classifiers can be bypassed with obfuscation and semantic tricks
Relevance enforcement	Built-in — rules can require replies to relate to the story or parent comment	Not possible — no concept of conversational structure
User experience	Instant, transparent feedback with explanations and suggestions that reduce frustration	Category‑based feedback only. Reasons are granular and cannot reference context or provide tailored guidance
Autonomous operation	Yes — ideal for self-moderating story boards and forums	No — requires server-side moderation infrastructure

Why Relevio is Hard to Fool?

Relevio isn’t just another AI moderation layer. It’s a rule‑driven, context‑aware system built to withstand the tricks, obfuscations, and evasions that routinely bypass classifier‑based moderation engines.

Semantic understanding beats surface‑level tricks

Relevio uses an LLM as a semantic interpreter, not a classifier. That means it can understand meaning even when users try to disguise it with misspellings, homoglyphs, emoji breaks, or paraphrasing. Classifier‑based systems can be fooled by surface‑level obfuscation; Relevio reads intent.

Deterministic rules prevent model drift

The LLM never decides the outcome. Your ruleset does.

Relevio uses the model only to interpret the content and context, then applies your rules deterministically. This eliminates the unpredictability and “mood swings” that make other AI moderation systems inconsistent.

Injection‑resistant by design

Relevio’s prompt architecture is tightly constrained:

• user content is treated strictly as data
• rules cannot be overridden
• injection attempts cause automatic failure
• malformed structures are rejected
• the output schema is fixed and enforced

This fail‑closed design makes instruction injection effectively impossible.

Context makes evasion harder

Every check includes the relevant story, comment, or reply chain.

Users can’t simply post something ambiguous or evasive — Relevio evaluates meaning in context, so relevance‑dodging and semantic misdirection don’t work.

Cryptographic signatures enforce integrity

Even if someone tried to tamper with content after moderation, the HMAC signature would break. Your backend only accepts content that matches its signature, making post‑moderation manipulation impossible.

Fail‑closed behaviour stops edge‑case exploits

If anything looks suspicious — malformed JSON, conflicting labels, unclear intent — Relevio returns a fail. Attackers don’t get partial credit or ambiguous outcomes to exploit.

The result

A moderation engine that’s far harder to bypass than classifier‑based systems and far more predictable than unconstrained LLM prompts. Relevio combines semantic understanding with deterministic rules and cryptographic trust, making it exceptionally resistant to adversarial phrasing, obfuscation, and manipulation.

What's the trade-off?

Relevio is slower and more expensive than classifier‑based moderation engines.

A classifier like Hive returns a score in well under a second and at a fraction of a cent because:

• it's a fixed‑architecture model
• it runs on a narrow set of categories
• it doesn’t interpret rules
• it doesn’t process long context
• it doesn’t generate explanations
• it doesn’t apply a cryptographic signature

Relevio, by contrast, is:

• performing full semantic analysis
• applying your deterministic rules
• checking context
• evaluating images and text together
• generating clear actionable feedback
• applying a cryptographic signature

That’s simply more work meaning it will take several seconds and it will cost orders of magnitude more.

Why the cost is an advantage

The extra cost buys something far more valuable. Because you eliminate the entire moderation infrastructure that a classifier based system requires. With Relevio, you don’t need:

• ingestion pipelines
• queues or workers
• dashboards
• human review
• appeals workflows
• storage for flagged content
• trust and safety staff
• backend scaling for moderation traffic

Your backend only verifies a fast HMAC. Everything else happens in the browser.

So while Relevio costs more per request, it costs far less overall to integrate and operate. You swap a fraction of a cent for the removal of an entire moderation department.

Why the delay works in Relevio’s favour

When moderation is client‑side, the interface can absorb the wait with:

• inline “checking your post…” animations
• optimistic UI that keeps the user in flow
• progressive hints and micro‑interactions
• a sense of active assistance rather than blocking

Users already expect this kind of rhythm from tools like spellcheckers, grammar assistants, and AI writing helpers. A short pause feels normal when the system is doing something meaningful.

The payoff

Relevio gives users something far more valuable than raw speed:

• specific explanations
• rule‑level clarity
• suggestions for improvement
• context‑aware reasoning
• multimodal consistency checks
• cryptographic trust in the result

Relevio’s a moderation engine that doesn’t just say “no” — it helps people write better content.

The result

A few seconds of latency buys you:

• richer moderation
• safer communities
• happier users
• a feather‑light backend
• a system that scales without moderation staff
• a platform that moderates itself

Relevio trades milliseconds and that fraction of a cent for a better business.