> ## Documentation Index
> Fetch the complete documentation index at: https://safety.flamy.net/llms.txt
> Use this file to discover all available pages before exploring further.

# Prohibited content

> Hard limits. What Flamify will not host under any circumstances.

# Prohibited content

> Hard limits. What Flamify will not host under any circumstances.

The categories below are rejected in chat, removed if discovered after the
fact, and — where the conduct is criminal — reported to the appropriate
authorities. There are no creative-framing exceptions and no “clearly
fictional” carve-outs for these hard limits.

If you see prohibited content on the platform, please
[report it](/reporting/how-to-report).

## Scope

Hard limits apply across the product: public catalog, private chats, and
private generations. [Safe filters](/moderation/safe-filters) add stricter
SFW rules on designated brand surfaces; they never loosen the limits below.

## Categories

### Minors / underage

Any content depicting, implying, or coded to suggest individuals under 18 in
a sexual, suggestive, or romantic context. This includes:

* Explicit age statements under 18.
* Settings or framings that imply a minor (schoolchild scenarios, “about to
  turn 18”, “very young looking”, loli-style cues).
* Visual cues that, in body proportions, read as a minor regardless of
  stated age.
* Grooming, sexualised training, age-play, or “ageing” mechanics used to
  sexualise a minor.

**Enforcement:** chat **STOP** (`#sexual_minors`); characters cannot be
created under 18; appearance negatives always include child / underage /
teen blockers.

Content involving minors is removed and, where the conduct is criminal,
reported to the appropriate national child-safety body (including NCMEC
CyberTipline in the United States and equivalent bodies in other
jurisdictions).

### Self-harm and suicide

We do not generate or continue conversations that encourage, instruct, or
romanticise suicide or self-harm.

**Enforcement:** chat **STOP** (`#selfharm`) with a crisis-oriented fixed
response. Genuine help-seeking language is redirected toward professional
resources; the AI companion is not a substitute for care.

### Incest / biological family sexual content

Sexual or romantic content implying a biological family relationship.

**Enforcement:** chat **STOP** (`#incest`).

### Bestiality / zoophilia

Sexual content involving non-anthropomorphic animals.

**Enforcement:** chat **STOP** (`#zoophilia`). Anthropomorphic characters
with clearly adult human features remain subject to the rest of this policy.

### Extreme gore and torture

Gratuitous depictions of serious injury, mutilation, torture, or death as
the appeal of the scene.

**Enforcement:** chat **STOP** (`#gore`). Narrative conflict in written
roleplay is evaluated in context by the verifier; shock-for-shock’s-sake
violence is not allowed.

### Real-world crime facilitation

Content that plans, instructs, solicits, or provides real-world facilitation
for crime, including:

* Weapons / explosives manufacture (`#weapon` — **STOP**)
* Illicit drug manufacture or sourcing instructions (`#drugs` — **STOP**)
* Extortion / blackmail (`#extortion` — **STOP**)
* Grooming / coercive manipulation, especially hiding contact from parents
  or adults (`#manipulation` — **STOP**)

### Eating-disorder promotion

Content that encourages or instructs disordered eating as aspirational
identity (“pro-ana” and equivalents).

**Enforcement:** chat **HOLD** (`#eating_disorder`) with a redirect toward
professional help.

### Therapy substitution

Requests that the AI act as a licensed therapist / psychiatrist / counselor
as a replacement for real care.

**Enforcement:** chat **HOLD** (`#therapy`). Flamify is not a clinical
service.

### Real-life meetings

Attempts to move the interaction into real-world meetings, addresses, or
in-person contact with the AI persona.

**Enforcement:** chat **HOLD** (`#irl_meeting`). Characters are software;
they do not meet users offline.

### Identity confusion

Attempts to treat the AI as a real human in a way that obscures the
product’s nature.

**Enforcement:** chat **HOLD** (`#identity_confusion`) with a clear
disclosure that the assistant is virtual.

## What is allowed (contrast)

Within the hard limits above, the main adult product allows:

* Original adult characters (18+) of any gender, body type, or ethnicity.
* Consensual adult erotic and romantic roleplay between fictional adults.
* Fantasy, sci-fi, supernatural, and historical settings.
* Power-exchange / BDSM language between fictional adults when framed as
  adult roleplay (the verifier is instructed not to confuse this with
  grooming or non-consensual real-world harm).

Safe Mode brand surfaces additionally restrict NSFW discovery and generation
— see [Safe filters](/moderation/safe-filters).

## Evasion

Hiding any of the above via encoding, obfuscation, language switching, or
repeated near-duplicate attempts is treated as the underlying violation plus
a separate abuse of the platform.

## Edge cases

Ambiguous cases are escalated to human review. For trust and policy
questions not covered here, see [Contact](/contact).