mcp-rule34
This server lets a chat client read rule34.xxx: look up tag spellings, search posts by tags, and read full post records.
find_tags — Check how rule34.xxx spells a tag and how many posts carry it, so you can search with the site's exact tag names.
search_posts — Find posts by tags, combining required tags, alternatives (
any_of), and exclusions (exclude).Filter search results — By media type (image/animated/video), rating (questionable/explicit), sort order (score/id/updated/random), page, and limit.
Get search totals — See how many posts the whole search matches, not just the returned page.
get_post — Read one post's full record by its numeric id or its rule34.xxx page URL.
Rich post metadata — Each post can include page/file/preview/sample URLs, dimensions, score, rating, tag list, uploader, comment count, source, dates, parent, and more.
Requires credentials — A rule34.xxx account and personal API key (RULE34_USER_ID and RULE34_API_KEY) must be provided.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mcp-rule34find rule34 posts tagged 'sonic' and 'amy', rated safe"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
mcp-rule34
rule34.xxx is a large image board whose posts are indexed entirely by tags, and it publishes an API for reading them. A post carries its identifier, the address of its page and of its file, its dimensions, the score its viewers gave it, its rating, and the full list of tags it was filed under. The site holds tens of millions of posts and answers a tag search with the number it matched.
This server connects a chat client to that index. You can search the posts by tags, combining required tags, alternatives and exclusions, read one post's record by its identifier, and look up how the site spells a tag before searching on it. It needs an account and an API key, which the site issues per person.
Install
Claude Code
claude mcp add rule34 --env RULE34_USER_ID=your-id --env RULE34_API_KEY=your-key -- npx -y mcp-rule34Claude Desktop, Cursor, and any client using the standard config format
{
"mcpServers": {
"rule34": {
"command": "npx",
"args": ["-y", "mcp-rule34"],
"env": {
"RULE34_USER_ID": "your-id",
"RULE34_API_KEY": "your-key"
}
}
}
}Node 24 or later is required. The two credentials are required; everything else under Configuration is optional.
With Docker
{
"mcpServers": {
"rule34": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"RULE34_USER_ID",
"-e",
"RULE34_API_KEY",
"ghcr.io/smeet666/mcp-rule34:2.0.2"
]
}
}
}-i keeps stdin open, which is where the protocol travels, and -t is left out
because a TTY rewrites the stream. The container needs outbound HTTPS to
api.rule34.xxx and rule34.xxx, and the two credentials from your environment:
no volume, no port.
Bundle, without npm
Download mcp-rule34-2.0.2.mcpb from
the latest release and
open it. A client that supports MCP bundles installs it on its own, with no npm
to run. The credentials are still set in the client's configuration.
Related MCP server: Civitai MCP Server
What you can ask
"How does the site spell that tag?"
"Find posts tagged with those two, sorted by score."
"Same search, leaving out monochrome."
"Read me the record for post 1234567."
"How many posts does that search match?"
The ordinary path runs from a tag lookup to a search: find_tags gives the
spelling the site uses, and search_posts takes it.
Tools
Tool | What it does |
| Finds how the site spells a tag, and how many posts carry it. |
| Searches the posts by tags, with alternatives and exclusions. |
| Reads one post's record by its identifier or its address. |
Look a tag up before searching on it. The site indexes by tag alone, and a tag it does not hold matches nothing, which reads as an empty result rather than as a misspelling.
find_tags
Finds how the site spells a tag.
Argument | Type | Required | What it does |
| string, up to 60 characters | yes | The text to look up. |
In return: query as it was sent, and tags, each with the spelling the site
uses and the number of posts carrying it.
search_posts
Searches the posts by tags.
Argument | Type | Required | What it does |
| array of 1 to 10 tag names | yes | Tags a post must carry. |
| array of 1 to 10 tag names | no | Tags a post must carry at least one of. |
| array of 1 to 10 tag names | no | Tags a post must carry none of. |
|
| no | Stills, GIFs or MP4. |
|
| no | Both are searched when this is left out. |
|
| no | The order the site sorts by. |
| integer, 1 to 100, default | no | Posts to serve. |
| integer, 1 to 200, default | no | Which page of posts. |
media_type is read off the site's own tags, so the classification is as good as
the tagging. The two ratings above are the only ones the site holds: any other
value answers zero posts and no error, which would hand back an absence the site
never had.
In return: tags, any_of and exclude as the site spells them; query,
the search as it was sent in the site's own language; total, the posts the
whole search matches as counted by the site; and the posts themselves, each with
its id, post_url, file_url, preview_url, sample_url, width,
height, score, rating and tags. The score is the site's own and it is
updated once a day.
get_post
Reads one post's record.
Argument | Type | Required | What it does |
| integer, 1 or more | one of two | The post identifier. |
| a rule34.xxx post URL | one of two | The address of the post's page. |
In return: the post a search row carries, with its full tag list.
Configuration
The two credentials are required. Everything else is optional, and all of it goes
in the env block of your client config.
Variable | Default | What it does |
| none, required | Your numeric account id. |
| none, required | Your API key, which is personal. |
| the project identity | Names your application to the site, with an address where a person can be reached. |
|
| Gap between two requests, from 1000 to 60000. |
|
| Deadline for one request, from 1000 to 120000. |
|
| Attempts after a transient failure, from 0 to 10. |
|
| How long an answer stays in memory, from 0 to 86400000. |
|
| Answers held in memory at once, from 0 to 10000. |
|
|
|
Where the credentials come from. Sign in, open
Account options, and find
the row named API Access Credentials. It shows &api_key=…&user_id=…, and
those two values are what goes in the configuration above. If the key is empty,
tick Generate New Key? and save.
The site issues one key per person, and asks that applications serving its content display no advertising and put it behind no paywall. This server ships no key of its own, and each user brings their own. Started without credentials, it runs, publishes its tools, and answers every call by naming the two variables to set: it sends no request it knows the site will refuse.
A value outside its range falls back to the default, and the reason is written to stderr.
Errors
Every failure carries one of six codes, a message, and where it helps a hint naming the next move.
Code | What happened | What to do |
| The site answered, and holds nothing at that address. | Check the identifier with |
| The arguments were refused before any request went out. | Read the message, which names the argument. A missing credential is reported here. |
| The site asked this client to slow down. | Wait the number of seconds the hint names and call again with the same arguments. The post is still there. |
| The answer arrived in a shape this client cannot read. | Report it at the issue tracker. |
| The request did not complete. | Try again shortly. |
| The request passed its deadline. | Raise |
As a library
The layer reading the site is published on its own, with its pacing, its cache and its errors, and with no protocol attached.
import { Rule34Client } from "mcp-rule34/client";
const client = new Rule34Client({ credentials: { userId: "…", apiKey: "…" } });
const { data, cached } = await client.searchPosts({ tags: ["example"], limit: 5 });
console.log(data.total, cached);Each read answers { data, cached }, and throws an error carrying one of the six
codes. The one-second floor between two requests holds here as well.
Pacing and attribution
Requests go out one at a time with at least a second between them, and that floor
holds however the server is configured. The User-Agent always ends with the
project identity and an address where a person can be reached.
Reads go to the API the site documents, and the credentials it issues are what its terms ask automated reading to use. Every result carries the address of the post's page. Nothing is downloaded: a file address travels through an answer as a string.
This MCP server is an unofficial project, with no affiliation to rule34.xxx.
Privacy
This server collects nothing about you and sends nothing to its author. It runs
on your machine, contacts api.rule34.xxx and rule34.xxx and nothing else,
holds its answers in memory while it runs, and writes nothing to disk. Your
credentials are read from the environment and sent to that site alone.
PRIVACY.md states what a request carries and which settings change
any of it.
Development
npm install
npm run build:fixtures
npm test
npm run checkTests run against generated fixtures and make no network request. The live suite,
npm run test:live, makes one request per route and runs nightly against the
site itself.
Contributing
Bugs, questions and ideas belong in the issue tracker. Pull requests are welcome; opening an issue first helps agree on the shape of the change. See CONTRIBUTING.md.
License
MIT, see LICENSE. The posts and the tags belong to rule34.xxx and to the people who uploaded them.
mcp-rule34 (français)
rule34.xxx est un grand imageboard dont les publications sont indexées entièrement par étiquettes, et il publie une API pour les lire. Une publication porte son identifiant, l'adresse de sa page et de son fichier, ses dimensions, la note que ses visiteurs lui ont donnée, son classement, et la liste complète des étiquettes sous lesquelles elle est rangée. Le site contient des dizaines de millions de publications et répond à une recherche par étiquettes en donnant le nombre qu'elle a trouvé.
Ce serveur relie un client de conversation à cet index. On peut chercher les publications par étiquettes, en combinant les étiquettes exigées, les alternatives et les exclusions, lire la fiche d'une publication par son identifiant, et vérifier comment le site orthographie une étiquette avant de chercher dessus. Il demande un compte et une clé d'API, que le site délivre par personne.
Installation
Claude Code
claude mcp add rule34 --env RULE34_USER_ID=votre-id --env RULE34_API_KEY=votre-cle -- npx -y mcp-rule34Claude Desktop, Cursor, et tout client au format de configuration standard
{
"mcpServers": {
"rule34": {
"command": "npx",
"args": ["-y", "mcp-rule34"],
"env": {
"RULE34_USER_ID": "votre-id",
"RULE34_API_KEY": "votre-cle"
}
}
}
}Node 24 ou plus récent est nécessaire. Les deux identifiants sont obligatoires ; tout le reste, sous Configuration, est facultatif.
Avec Docker
{
"mcpServers": {
"rule34": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"RULE34_USER_ID",
"-e",
"RULE34_API_KEY",
"ghcr.io/smeet666/mcp-rule34:2.0.2"
]
}
}
}-i garde l'entrée standard ouverte, qui est le canal du protocole, et -t est
omis parce qu'un TTY réécrit le flux. Le conteneur a besoin d'un accès HTTPS
sortant vers api.rule34.xxx et rule34.xxx, et des deux identifiants pris dans
votre environnement : aucun volume, aucun port.
Bundle, sans npm
Téléchargez mcp-rule34-2.0.2.mcpb depuis
la dernière publication
et ouvrez-le. Un client qui gère les bundles MCP l'installe seul, sans npm à
lancer. Les identifiants se posent toujours dans la configuration du client.
Ce qu'on peut demander
« Comment le site orthographie-t-il cette étiquette ? »
« Trouve les publications portant ces deux étiquettes, triées par note. »
« La même recherche, sans monochrome. »
« Lis-moi la fiche de la publication 1234567. »
« Combien de publications cette recherche trouve-t-elle ? »
Le chemin ordinaire va d'une recherche d'étiquette à une recherche de
publications : find_tags donne l'orthographe qu'emploie le site, et
search_posts la reprend.
Les outils
Outil | Ce qu'il fait |
| Trouve l'orthographe d'une étiquette, et combien de publications la portent. |
| Cherche les publications par étiquettes, avec alternatives et exclusions. |
| Lit la fiche d'une publication par son identifiant ou son adresse. |
Vérifiez une étiquette avant de chercher dessus. Le site indexe par étiquette seule, et une étiquette qu'il ne connaît pas ne correspond à rien, ce qui se lit comme un résultat vide plutôt que comme une faute d'orthographe.
find_tags
Trouve l'orthographe d'une étiquette.
Argument | Type | Requis | Ce qu'il fait |
| chaîne, jusqu'à 60 caractères | oui | Le texte à vérifier. |
En retour : query tel qu'envoyé, et tags, chacune avec l'orthographe
qu'emploie le site et le nombre de publications qui la portent.
search_posts
Cherche les publications par étiquettes.
Argument | Type | Requis | Ce qu'il fait |
| tableau de 1 à 10 noms d'étiquettes | oui | Étiquettes qu'une publication doit porter. |
| tableau de 1 à 10 noms d'étiquettes | non | Étiquettes dont une publication doit porter au moins une. |
| tableau de 1 à 10 noms d'étiquettes | non | Étiquettes qu'une publication ne doit pas porter. |
|
| non | Images fixes, GIF ou MP4. |
|
| non | Les deux sont cherchés quand il est omis. |
|
| non | L'ordre de tri du site. |
| entier, 1 à 100, défaut | non | Publications à servir. |
| entier, 1 à 200, défaut | non | Quelle page de publications. |
media_type se lit sur les étiquettes du site lui-même, donc la classification
vaut ce que vaut l'étiquetage. Les deux classements ci-dessus sont les seuls que
le site connaisse : toute autre valeur rend zéro publication et aucune erreur, ce
qui livrerait une absence que le site n'a jamais eue.
En retour : tags, any_of et exclude tels que le site les
orthographie ; query, la recherche telle qu'elle a été envoyée dans la langue
du site ; total, les publications que toute la recherche trouve, tel que le
site les compte ; et les publications elles-mêmes, chacune avec son id,
post_url, file_url, preview_url, sample_url, width, height, score,
rating et tags. La note est celle du site et elle est mise à jour une fois
par jour.
get_post
Lit la fiche d'une publication.
Argument | Type | Requis | Ce qu'il fait |
| entier, 1 ou plus | l'un des deux | L'identifiant de la publication. |
| une adresse de page rule34.xxx | l'un des deux | L'adresse de la page. |
En retour : la publication que porte une ligne de recherche, avec sa liste complète d'étiquettes.
Configuration
Les deux identifiants sont obligatoires. Tout le reste est facultatif, et tout se
pose dans le bloc env de la configuration du client.
Variable | Défaut | Ce qu'elle fait |
| aucun, obligatoire | L'identifiant numérique de votre compte. |
| aucun, obligatoire | Votre clé d'API, qui est personnelle. |
| l'identité du projet | Nomme votre application auprès du site, avec une adresse où joindre une personne. |
|
| Écart entre deux requêtes, de 1000 à 60000. |
|
| Délai d'une requête, de 1000 à 120000. |
|
| Tentatives après un échec passager, de 0 à 10. |
|
| Durée pendant laquelle une réponse reste en mémoire, de 0 à 86400000. |
|
| Réponses gardées en mémoire à la fois, de 0 à 10000. |
|
|
|
D'où viennent les identifiants. Connectez-vous, ouvrez
Account options, et
trouvez la ligne API Access Credentials. Elle affiche &api_key=…&user_id=…,
et ces deux valeurs sont ce qui va dans la configuration ci-dessus. Si la clé est
vide, cochez Generate New Key? et enregistrez.
Le site délivre une clé par personne, et demande que les applications servant son contenu n'affichent aucune publicité et ne le placent derrière aucun péage. Ce serveur n'embarque aucune clé, et chacun apporte la sienne. Démarré sans identifiants, il tourne, publie ses outils, et répond à chaque appel en nommant les deux variables à poser : il n'envoie aucune requête dont il sait que le site la refusera.
Une valeur hors de sa plage retombe sur le défaut, et la raison est écrite sur la sortie d'erreur.
Erreurs
Chaque échec porte un des six codes, un message, et quand cela aide une indication du geste suivant.
Code | Ce qui s'est passé | Que faire |
| Le site a répondu, et n'a rien à cette adresse. | Vérifiez l'identifiant avec |
| Les arguments ont été refusés avant toute requête. | Lisez le message, qui nomme l'argument. Un identifiant manquant est signalé ici. |
| Le site demande à ce client de ralentir. | Attendez les secondes indiquées et rappelez avec les mêmes arguments. La publication est toujours là. |
| La réponse est arrivée dans une forme illisible ici. | Signalez-le sur le suivi d'incidents. |
| La requête n'a pas abouti. | Réessayez sous peu. |
| La requête a dépassé son délai. | Augmentez |
Comme bibliothèque
La couche qui lit le site est publiée seule, avec son rythme, son cache et ses erreurs, sans protocole attaché.
import { Rule34Client } from "mcp-rule34/client";
const client = new Rule34Client({ credentials: { userId: "…", apiKey: "…" } });
const { data, cached } = await client.searchPosts({ tags: ["example"], limit: 5 });
console.log(data.total, cached);Chaque lecture répond { data, cached }, et lève une erreur portant un des six
codes. Le plancher d'une seconde entre deux requêtes tient également ici.
Rythme et attribution
Les requêtes partent une à une avec au moins une seconde entre elles, et ce
plancher tient quelle que soit la configuration. Le User-Agent se termine
toujours par l'identité du projet et une adresse où joindre une personne.
Les lectures passent par l'API que le site documente, et les identifiants qu'il délivre sont ce que ses conditions demandent d'employer pour une lecture automatisée. Chaque résultat porte l'adresse de la page de la publication. Rien n'est téléchargé : une adresse de fichier traverse une réponse comme une chaîne de caractères.
Ce MCP est un projet non officiel, sans affiliation à rule34.xxx.
Confidentialité
Ce serveur ne collecte rien sur vous et n'envoie rien à son auteur. Il tourne sur
votre machine, ne joint que api.rule34.xxx et rule34.xxx, garde ses réponses
en mémoire le temps qu'il tourne, et n'écrit rien sur le disque. Vos identifiants
sont lus dans l'environnement et envoyés à ce seul site. PRIVACY.md
dit ce qu'une requête emporte et quels réglages changent cela.
Développement
npm install
npm run build:fixtures
npm test
npm run checkLes tests s'exécutent sur des fixtures engendrées et n'émettent aucune requête.
La suite en direct, npm run test:live, émet une requête par route et tourne
chaque nuit contre le site lui-même.
Contribuer
Les anomalies, les questions et les idées ont leur place dans le suivi d'incidents. Les propositions de modification sont bienvenues ; ouvrir un ticket d'abord aide à s'accorder sur la forme du changement. Voir CONTRIBUTING.md.
Licence
MIT, voir LICENSE. Les publications et les étiquettes appartiennent à rule34.xxx et aux personnes qui les ont déposées.
Available Tools
3 toolsfind_tagsFind how a tag is spelledARead-onlyIdempotent
Find how rule34.xxx spells a tag, and how many posts carry it. The site matches from the start of a name, so 'kimagure' finds 'kimagure_orange_road' while 'orange road' finds nothing. It offers at most ten names. Use this before searching when a name might be spelled differently, and after a search that found nothing because a tag does not exist.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | The start of a tag name, such as 'kimagure' or 'prince of tennis'. Spaces are joined with underscores. |
Output Schema
| Name | Required | Description |
|---|---|---|
| tags | Yes | |
| notes | Yes | |
| query | Yes | The text as rule34.xxx would spell it. |
| source | Yes |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false. The description adds valuable behavior beyond those: prefix matching ('matches from the start of a name'), the ten-result limit, and that it returns the spelling and post count. No contradiction with annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three sentences, no fluff. The main purpose is front-loaded, followed by a concrete matching example, then usage guidance. Every sentence serves a distinct role and the description is appropriately sized for the tool's simplicity.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With an output schema present (so return values are documented elsewhere) and the single parameter fully described in the schema, the description covers all remaining contextual needs: when to use, how matching works, result limit, and how it relates to searching. Nothing an agent needs to call it correctly is missing.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema's 'query' property description already fully covers the parameter semantics: 'The start of a tag name... Spaces are joined with underscores.' The tool description provides an illustrative example ('orange road' finds nothing) but does not add new parameter-level meaning beyond the schema, which has 100% coverage. Baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific verb ('Find') and precise resource ('how rule34.xxx spells a tag, and how many posts carry it'). It clearly differentiates from sibling tools (search_posts, get_post) by focusing on tag spelling and post count rather than post content or retrieval.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Explicitly instructs when to use: 'Use this before searching when a name might be spelled differently, and after a search that found nothing because a tag does not exist.' This gives clear context and ties to search_posts, and also notes the 'at most ten names' limitation.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
get_postRead one postARead-onlyIdempotent
Read one rule34.xxx post, named by its id or by a link to its page. Returns every tag the post carries, each with its kind (character, copyright, artist, general or metadata) and how many posts share it, alongside the image or video, the dimensions, the score, the rating, the uploader, the number of comments, the credited source and the date the site took the post in. Use this after a search, whose rows show only the first few tags of each post.
| Name | Required | Description | Default |
|---|---|---|---|
| id | No | The post's numeric id, as a search returns it. | |
| url | No | A link to the post's page, such as https://rule34.xxx/index.php?page=post&s=view&id=2195419. The link is read rather than followed. |
Output Schema
| Name | Required | Description |
|---|---|---|
| id | Yes | |
| md5 | Yes | |
| tags | Yes | |
| notes | Yes | |
| owner | Yes | The account that uploaded the post. |
| score | Yes | The site's own score, updated once a day. |
| width | Yes | |
| height | Yes | |
| rating | Yes | |
| source | Yes | |
| status | Yes | |
| file_url | Yes | |
| post_url | Yes | |
| has_notes | Yes | |
| parent_id | Yes | |
| changed_at | Yes | When the post last changed, ISO 8601. |
| created_at | Yes | When the post was added to rule34.xxx, ISO 8601. The site imported much of its older catalogue in bulk, so thousands of posts share one day. |
| sample_url | Yes | |
| preview_url | Yes | |
| source_site | Yes | |
| comment_count | Yes |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond the readOnlyHint and idempotentHint annotations, the description discloses detailed behavior: it returns every tag with its kind and post count, media type, dimensions, score, rating, uploader, comment count, source, and ingestion date. It also clarifies that a URL link is 'read rather than followed,' which is useful behavioral nuance not present in the annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is front-loaded with the core action and identifier modes, then lists returned data, then gives usage context. It is a single long sentence with a dense enumeration, but each element carries information and nothing is filler. Minor redundancy with the title and output schema keeps it from a 5.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the rich output schema, strong annotations, and sibling context, the description covers what the agent needs to know: when to invoke (after a search), how to identify the post (id or URL), and what behavior to expect (read-only, URL not followed). No critical decision-making information is missing.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema already documents both parameters well (100% coverage), but the description adds meaning by framing them as equivalent alternatives: 'named by its id or by a link to its page.' This compensates for the schema's lack of an explicit oneOf constraint, though the description could have been even more explicit about exactly-one semantics.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific verb ('Read'), a single resource ('one rule34.xxx post'), and the two ways to identify it ('id or link'), which clearly distinguishes it from the sibling search_posts and find_tags tools. It also names exactly what the tool returns, so there is no ambiguity about its function.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly says 'Use this after a search, whose rows show only the first few tags of each post,' giving clear when-to-use context versus search_posts. It does not explicitly enumerate when not to use it or mention find_tags as an alternative, but the single vs. search distinction is reasonably clear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
search_postsSearch posts by tagARead-onlyIdempotent
Search rule34.xxx for posts carrying a tag, or several tags at once. Tags are written as one token, so a name said as words is joined by underscores: 'asuka langley sohryu' and 'ranma 1/2' are single tags, and this tool joins them for you. Returns each post's page, its image or video URL, dimensions, score and rating, alongside the number of posts the whole search matches. A row shows the first 12 of a post's tags and says how many it has; call get_post for the whole list with a kind for every tag. Use 'media_type' to ask for still images, animations or video; the site classifies these itself. When a search finds nothing, the answer names the tags rule34.xxx does not hold, or says how many posts the tags match once the restrictions are dropped.
| Name | Required | Description | Default |
|---|---|---|---|
| page | No | ||
| sort | No | 'score' is the site's own tally of votes, recomputed once a day. 'id' is newest first. 'updated' is by when a post last changed, which retagging moves. 'random' draws a different set on each call. | score |
| tags | Yes | Tags a post must all carry, such as ['asuka langley sohryu', 'black hair']. | |
| limit | No | ||
| any_of | No | Tags a post must carry at least one of. | |
| rating | No | The two ratings rule34.xxx holds. Leave unset to search both. | |
| exclude | No | Tags a post must not carry, such as ['monochrome'] to leave out black and white. | |
| media_type | No | 'image' for stills, 'animated' for GIFs, 'video' for MP4. The site states this in its own tags, so the classification is as good as the tagging. | any |
Output Schema
| Name | Required | Description |
|---|---|---|
| page | Yes | |
| tags | Yes | The required tags, as rule34.xxx spells them. |
| notes | Yes | |
| posts | Yes | |
| query | Yes | The search as it was sent, in the site's own language. |
| total | Yes | Posts the whole search matches, as counted by the site. |
| any_of | Yes | The alternatives, as rule34.xxx spells them. |
| source | Yes | |
| exclude | Yes | The excluded tags, as rule34.xxx spells them. |
| has_more | Yes | |
| next_page | Yes |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description discloses what each result row contains, how many tags are shown, total match count, and the exact behavior when no results are found. These details meaningfully exceed the annotations, which already mark the tool as read-only and idempotent.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is dense but every sentence earns its place: scope, tag formatting, return contents, sibling routing, media_type nuance, and empty-result behavior. It is front-loaded with the core action.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For an 8-parameter search tool, the description covers input formatting, output shape, pagination-relevant count, media filtering, and failure behavior. Since an output schema exists, nothing essential is missing for an agent to call the tool correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
With 75% schema coverage, the schema already documents most parameters. The description adds real value by explaining that multi-word tags are underscore-joined and that the tool performs the joining, and by clarifying that media_type relies on the site's own tagging.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description opens with a specific verb and resource: 'Search rule34.xxx for posts carrying a tag, or several tags at once.' It also distinguishes itself from the sibling get_post by telling the agent to call get_post for a post's full tag list.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It gives an explicit hand-off to get_post for the complete tag list, and explains when media_type should be used. It does not mention find_tags as an alternative, but the core search use case is unambiguous.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
2 tool updates
v2.0.1- Changed
get_post12 fields changed- removed
Output schema / properties / changed_at / anyOfRemoved value: -[ - { - "type": "string" - }, - { - "type": "null" - } -] - added
Output schema / properties / changed_at / typeAdded value: +[ + "string", + "null" +] - removed
Output schema / properties / created_at / anyOfRemoved value: -[ - { - "type": "string" - }, - { - "type": "null" - } -] - added
Output schema / properties / created_at / typeAdded value: +[ + "string", + "null" +] - removed
Output schema / properties / owner / anyOfRemoved value: -[ - { - "type": "string" - }, - { - "type": "null" - } -] - added
Output schema / properties / owner / typeAdded value: +[ + "string", + "null" +] - removed
Output schema / properties / preview_url / anyOfRemoved value: -[ - { - "type": "string" - }, - { - "type": "null" - } -] - added
Output schema / properties / preview_url / typeAdded value: +[ + "string", + "null" +] - removed
Output schema / properties / sample_url / anyOfRemoved value: -[ - { - "type": "string" - }, - { - "type": "null" - } -] - added
Output schema / properties / sample_url / typeAdded value: +[ + "string", + "null" +] - removed
Output schema / properties / source / anyOfRemoved value: -[ - { - "type": "string" - }, - { - "type": "null" - } -] - added
Output schema / properties / source / typeAdded value: +[ + "string", + "null" +]
- Changed
search_posts8 fields changed- removed
Output schema / properties / posts / items / properties / created_at / anyOfRemoved value: -[ - { - "type": "string" - }, - { - "type": "null" - } -] - added
Output schema / properties / posts / items / properties / created_at / typeAdded value: +[ + "string", + "null" +] - removed
Output schema / properties / posts / items / properties / preview_url / anyOfRemoved value: -[ - { - "type": "string" - }, - { - "type": "null" - } -] - added
Output schema / properties / posts / items / properties / preview_url / typeAdded value: +[ + "string", + "null" +] - removed
Output schema / properties / posts / items / properties / sample_url / anyOfRemoved value: -[ - { - "type": "string" - }, - { - "type": "null" - } -] - added
Output schema / properties / posts / items / properties / sample_url / typeAdded value: +[ + "string", + "null" +] - removed
Output schema / properties / posts / items / properties / source / anyOfRemoved value: -[ - { - "type": "string" - }, - { - "type": "null" - } -] - added
Output schema / properties / posts / items / properties / source / typeAdded value: +[ + "string", + "null" +]
3 tool updates
v1.1.0- First observed
find_tags - First observed
get_post - First observed
search_posts
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose: search_posts queries by tags, get_post retrieves one post's full details, and find_tags resolves tag spellings. There is no overlap or ambiguity between them.
All tool names follow the same verb_noun pattern: search_posts, get_post, find_tags. The naming is predictable and consistent, making it easy for an agent to infer what each tool does.
Three tools is an appropriate scope for a focused rule34.xxx client: find tags, search posts, and inspect a single post. Each tool earns its place and there is no unnecessary bloat.
The tool set covers the core user workflow end to end: discovering the correct tag, searching posts with filters, and retrieving complete post metadata. No obvious missing operation is needed for the server's stated purpose.
Maintenance
Related MCP Connectors
Search adult porn video (NSFW, 18+): nude clips, creators, hashtags. Read-only, no API key.
Search adult porn video (NSFW, 18+): nude clips, creators, hashtags. Read-only, no API key.
Search adult porn video (NSFW, 18+): nude clips, creators, hashtags. Read-only, no API key.
X / Twitter public post, comment, reply, user, and search tools.
Related MCP Servers
- FlicenseNot gradedqualityFmaintenanceEnables searching and analyzing Danbooru tags, character statistics, and wiki information. Extracts tags from posts, analyzes frequently used tags for characters with customizable filters, and retrieves detailed wiki entries.1-
- AlicenseNot gradedqualityDmaintenanceEnables interaction with the Civitai API to search for AI models, browse images, and access creator information. It provides tools for filtering models by popularity, rating, or type and retrieving detailed version metadata.MIT
- AlicenseAqualityDmaintenanceIntegrates ThePornDB API into MCP-compatible applications to search for adult video scenes, movies, and JAV content. It enables users to retrieve detailed performer profiles and comprehensive content metadata through specialized tools for LLM applications.1864MIT
- AlicenseAqualityCmaintenanceSearches Danbooru for anime/game characters and their associated tags, providing tools to retrieve character profiles, co-occurring tags, wiki descriptions, and tag implications.512MIT