Detect AI slop in text, HTML, pages, and project folders. Free, no API key.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Run the judgement checklist (--judge) over text, HTML, or a directory: the 5-question review (purpose/audience/decision/variety/defaults) with machine-flagged signals per question, a craft-floor gate (contrast/alignment/text-hierarchy/spacing/no-fake-data/no-decorative-effects) mapped from scanner evidence with human-checkbox items for the non-machineable parts, refusal scaffolds, and a machine-computed PASS | REVIEW | REFUSE verdict computed ONLY from machine-verifiable parts. Same input always gives the same verdict.
directory path to judge — local filesystem access only
HTML document to judge
plain text to judge
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http ai-slop-checker https://slop-checker.lazynext.com/mcp
Run the judgement checklist (--judge) over text, HTML, or a directory: the 5-question review (purpose/audience/decision/variety/defaults) with machine-flagged signals per question, a craft-floor gate (contrast/alignment/text-hierarchy/spacing/no-fake-data/no-decorative-effects) mapped from scanner evidence with human-checkbox items for the non-machineable parts, refusal scaffolds, and a machine-computed PASS | REVIEW | REFUSE verdict computed ONLY from machine-verifiable parts. Same input always gives the same verdict.
| Argument | Type | Description |
|---|---|---|
| dir | string | directory path to judge — local filesystem access only |
| html | string | HTML document to judge |
| text | string | plain text to judge |
{
"type": "object",
"oneOf": [
{
"required": [
"text"
]
},
{
"required": [
"html"
]
},
{
"required": [
"dir"
]
}
],
"properties": {
"dir": {
"type": "string",
"description": "directory path to judge — local filesystem access only"
},
"html": {
"type": "string",
"description": "HTML document to judge"
},
"text": {
"type": "string",
"description": "plain text to judge"
}
}
}List the full pattern corpus (categories + every rule with id, severity, kind, match, fix). Use this to know what the scanner flags.
No arguments.
{
"type": "object",
"properties": {}
}Apply conservative safe rewrites (whole-word synonym swaps and phrase deletions only) to a text. Returns the rewritten text plus the change list.
| Argument | Type | Description |
|---|---|---|
| text* | string | text to rewrite |
{
"type": "object",
"required": [
"text"
],
"properties": {
"text": {
"type": "string",
"description": "text to rewrite"
}
}
}Scan HTML markup for AI-slop writing AND visual design tells (gradients, kickers, glassmorphism, stat rows, ...).
| Argument | Type | Description |
|---|---|---|
| html* | string | HTML document to scan |
| profile | string | severity profile (default | strict | lenient) |
{
"type": "object",
"required": [
"html"
],
"properties": {
"html": {
"type": "string",
"description": "HTML document to scan"
},
"profile": {
"enum": [
"default",
"strict",
"lenient"
],
"type": "string",
"description": "severity profile (default | strict | lenient)"
}
}
}Scan a local directory tree (prose/markup/styles/code lenses) and return a pooled report: score, top offending files, category breakdown. Local filesystem access only.
| Argument | Type | Description |
|---|---|---|
| dir* | string | path to a directory |
| profile | string | severity profile (default | strict | lenient) |
| topFiles | integer | how many top files to report (default 10) |
{
"type": "object",
"required": [
"dir"
],
"properties": {
"dir": {
"type": "string",
"description": "path to a directory"
},
"profile": {
"enum": [
"default",
"strict",
"lenient"
],
"type": "string",
"description": "severity profile (default | strict | lenient)"
},
"topFiles": {
"type": "integer",
"description": "how many top files to report (default 10)"
}
}
}Run the rendered-tier design checks (r-*): kicker size, gradient text, tracked caps, glued arrows, marquee, hidden reveal, fold budget, screen word budget, section rhythm, fake progress, and WCAG contrast — all pure functions over a computed snapshot ({elements:[...]}, see src/render-checks.js). Pass an object snapshot; no browser needed.
| Argument | Type | Description |
|---|---|---|
| url | string | http(s) URL to collect then check (requires the optional playwright collector) |
| profile | string | severity profile (default | strict | lenient) |
| snapshot | object | computed snapshot: { viewport?, documentStyle?, elements: [...], screens? } |
{
"type": "object",
"properties": {
"url": {
"type": "string",
"description": "http(s) URL to collect then check (requires the optional playwright collector)"
},
"profile": {
"enum": [
"default",
"strict",
"lenient"
],
"type": "string",
"description": "severity profile (default | strict | lenient)"
},
"snapshot": {
"type": "object",
"description": "computed snapshot: { viewport?, documentStyle?, elements: [...], screens? }"
}
}
}Scan plain text for AI-slop writing tells. Returns score (0-100), verdict, per-category breakdown, and every match with a fix suggestion.
| Argument | Type | Description |
|---|---|---|
| text* | string | plain text to scan |
| profile | string | severity profile (default | strict | lenient) |
| rewrite | boolean | include safe auto-rewrites (text only) |
{
"type": "object",
"required": [
"text"
],
"properties": {
"text": {
"type": "string",
"description": "plain text to scan"
},
"profile": {
"enum": [
"default",
"strict",
"lenient"
],
"type": "string",
"description": "severity profile (default | strict | lenient)"
},
"rewrite": {
"type": "boolean",
"description": "include safe auto-rewrites (text only)"
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 21, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.