Autocomplete a tag fragment against the booru tag index, returning the canonical tag name with its post count and category. Use this to turn a natural word, partial fragment, or NON-ENGLISH term (Chinese/Japanese other_names like '星野', 'ホシノ') into the exact booru tag form before calling the search tools: user says → autocomplete query → pick returned name 'blue hair' → 'blue hair' → 'blue_hair' 'miku' → 'miku' → 'hatsune_miku' (or other) 'knight' → 'knight' → 'knight', 'female_knight', ... '星野'/'ホシノ' → '星野' → 'hoshino_(blue_archive)' (resolved via other_names + alias) 'amamya_kokoro' → 'amamya_kokoro' → 'amamiya_kokoro' (auto-corrected)

Behavior:

• On Danbooru (default) the query goes through the /autocomplete.json?type=tag_query endpoint, which performs ALIAS resolution, FUZZY/prefix matching, and matches against a tag's other_names (so Chinese/Japanese/Romaji inputs resolve to the English canonical tag). Auto-correction: if the input is a misspelling that resolves to a single canonical tag, the first result is the correct one (and is_alias will mark alias entries).

• You do NOT need a trailing '*'; the underlying endpoint treats the query as a tag-query prefix/fuzzy term.

• Multi-word fragments are allowed (pass verbatim, e.g. 'long skirt' or 'long_skirt').

• post_count lets you reject obscure tags (likely to return ~0 results in search_prompts). Prefer tags with post_count > 100.

• category: 0=general, 1=artist, 3=copyright, 4=character, 5=meta — use this to choose between, e.g. 'blue_archive' (3, copyright) vs an artist named 'blue_archive' (1).

• is_alias (bool) marks results that are alias entries; the name field already holds the canonical (non-alias) tag name to use.

• For AIBooru (provider='aibooru') the Danbooru autocomplete index is used (since AIBooru mirrors its tag database); for rule34/gelbooru/e621 search use search_prompts/search_posts once you've found the tag name here (booru tags are largely cross-compatible).

Clean an already-known set of booru tags into a ready-to-use prompt. Use this when you already have a tag list (e.g. user pasted one or you received raw tags from search_posts) and only want the cleaned output. No network request is made.

tag_string accepts the SAME raw booru format that search_posts returns in raw_tags: space-separated, multi-word tags joined by UNDERSCORE ('1girl long_hair blue_eyes smile'). It also accepts a comma-separated list of already-cleaned tags ('1girl, long hair, blue eyes, smile') for idempotent re-cleaning. Do NOT pass natural-language sentences — split them into booru tags first (use autocomplete_tags to find the canonical form of each).

RECOMMENDED FIRST CHOICE for any character/tag lookup.

Returns everything in one call: visual traits (co-occurrence frequencies), wiki description, multilingual aliases, and tag implications in both directions (what this tag implies, and what tags imply this one — e.g. all costume variants).

Aggregates these data sources in parallel:

1. Related tags (co-occurrence frequencies) -> visual traits.

2. Wiki page -> textual description and multilingual aliases.

3. Tag implications (antecedent) -> tags auto-added by this tag.

4. Tag implications (consequent) -> tags that auto-add this tag (e.g. costume variants like hoshino_(swimsuit)_(blue_archive) imply hoshino_(blue_archive)).

Each source degrades gracefully: a missing wiki page or empty implications do not fail the whole request.

When auto_resolve is enabled (default) and the queried tag does not exist on Danbooru, the tool corrects the tag name via autocomplete and rebuilds the profile. The correction is reported in resolved_from.

Args: tag (str): Character tag, e.g. 'hoshino_(blue_archive)'. Spaces are automatically converted to underscores; non-English queries (Chinese/Japanese other_names like '星野'/'ホシノ') are resolved via the autocomplete index. limit (int): Max characteristic tags (1-100, default 25). response_format (str): 'markdown' or 'json'. auto_resolve (bool): Auto-correct misspelled/non-canonical tags (default True).

Returns: str: Markdown profile or JSON. Error: "Error: ".

Examples: - "Give me everything about Hoshino from Blue Archive" -> tag='hoshino_(blue_archive)'. - "Full profile of Rem from Re:Zero" -> tag='rem_(re:zero)'.

Get the implication chain for a tag (A -> B auto-adds).

For character costume variants, use danbooru_get_character_profile which queries both directions. This tool queries only the antecedent direction (what does this tag imply).

A Danbooru implication A -> B means every post tagged A is automatically also tagged B. This tool queries the antecedent direction (what does this tag imply). For the reverse direction (what tags imply this tag, e.g. costume variants), use danbooru_get_character_profile which queries both directions.

For characters this often encodes the source work (hoshino_(blue_archive) -> blue_archive) and structural traits. Not every character tag has implications, so an empty list is a valid result.

Args: tag (str): Antecedent tag, e.g. 'hoshino_(blue_archive)'. Spaces are automatically converted to underscores. limit (int): Max implications (1-1000, default 50). response_format (str): 'markdown' or 'json'.

Returns: str: Markdown table or JSON. Error: "Error: ".

Examples: - "What tags are auto-added when tagging Hoshino?" -> tag='hoshino_(blue_archive)'. - "Trace the implication chain of pink_hair" -> tag='pink_hair'.

Get the textual wiki description for a tag.

For complete character info, use danbooru_get_character_profile instead. This tool retrieves the wiki page only. Wiki pages contain the canonical description, often listing a character's appearance, source work, voice actor, and alternate costumes.

The body is written in Danbooru DText markup; this tool converts it to readable plain text in markdown mode.

Args: title (str): Wiki title, e.g. 'hoshino_(blue_archive)'. Spaces are automatically converted to underscores. body_limit (int): Truncate body to N chars (0 = no limit). Useful because some wiki pages are very long. response_format (str): 'markdown' or 'json'.

Returns: str: Markdown or JSON. A 404 returns "Error: Resource not found. ...". Error: "Error: " otherwise.

Examples: - "Describe Hoshino from Blue Archive" -> title='hoshino_(blue_archive)'. - "What does the ahoge tag mean?" -> title='ahoge'.

Utility: find or verify a tag's exact canonical name.

NOT for character analysis — use danbooru_get_character_profile or danbooru_search_character for that. This tool is for finding/verifying tag names and listing related tag variants (e.g. all costume variants of a character).

Uses Danbooru's autocomplete endpoint as the primary search engine because it reliably resolves aliases (e.g. amamya_kokoro -> amamiya_kokoro) and performs prefix matching without requiring explicit * wildcards. The tags.json endpoint is used only as a fallback when autocomplete returns nothing.

Args: query (str): Name pattern. Wildcards (*) supported but often unnecessary — autocomplete does prefix/fuzzy matching, and also matches against the tag's other_names (so Chinese/Japanese queries resolve to the English canonical tag). category (str|None): Filter to one category: 'general', 'artist', 'copyright', 'character', or 'meta'. Omit for all categories. limit (int): Max results (1-200, default 25). order (str): 'count' (post count desc, default), 'name' (name asc), or 'date' (created desc). hide_empty (bool): Hide zero-post tags (default True). response_format (str): 'markdown' or 'json'. auto_resolve (bool): Fall back to tags.json if autocomplete returns nothing (default True).

Returns: str: Markdown table or JSON. Error: "Error: ".

Examples: - "Does the tag amamiya_kokoro exist?" -> query='amamiya_kokoro'. - "Find Amamiya Kokoro even if misspelled" -> query='amamya_kokoro' (autocomplete corrects it to amamiya_kokoro). - "List all Blue Archive character tags" -> query='*_(blue_archive)', category='character', order='count'.

Get visual trait frequencies for a character tag.

Best tool for character visual trait analysis. Use danbooru_get_character_profile if you also need wiki/implications. Given a character or copyright tag (e.g. hoshino_(blue_archive)), it returns the tags that most frequently appear alongside it on Danbooru posts. A tag with frequency 0.92 (like ahoge or pink_hair for Hoshino) appears in 92% of posts carrying the query tag, so it is a defining visual trait.

Results are ordered by co-occurrence frequency (descending). Meta tags such as highres are excluded by default because they describe image quality, not the character.

When auto_resolve is enabled (default) and the queried tag does not exist on Danbooru, the tool automatically queries the autocomplete endpoint to find the correct name (e.g. correcting a misspelled amamya_kokoro to amamiya_kokoro), re-runs the search with the corrected name, and reports the correction in resolved_from.

Args: tag (str): Canonical Danbooru tag, e.g. 'hoshino_(blue_archive)'. Spaces -> underscores automatically; non-English queries resolve via the autocomplete index when auto_resolve is on. limit (int): Max related tags to return (1-100, default 25). category (str|None): Filter to one category: 'general', 'artist', 'copyright', 'character', or 'meta'. Omit for all categories. min_frequency (float): Min co-occurrence frequency 0-1. A tag with frequency 0.9 appears in 90% of posts that carry the query tag. exclude_meta (bool): Drop meta tags (default True). response_format (str): 'markdown' or 'json'. auto_resolve (bool): Auto-correct misspelled tags (default True).

Returns: str: Markdown table or JSON. Error: "Error: ".

Examples: - "What does Hoshino from Blue Archive look like?" -> tag='hoshino_(blue_archive)' -> returns ahoge, pink_hair, blue_eyes, heterochromia, halo, ... - "Find all copyright tags related to this character" -> tag='hoshino_(blue_archive)', category='copyright'.

Search a booru and return the raw post list (no cleaning). The same search semantics as search_prompts apply:

• Query with one or more booru TAGs (underscores for multi-word: 'blue_hair', 'hatsune_miku'), never write natural language.

• Tags AND together — every tag must be present in each returned post. PREFER A SINGLE TAG; stacking tags usually returns 0 posts. The cleaned prompt is where 'more detail' lives, not in the search query.

• MULTI-TAG SEARCH IS SUPPORTED but each provider caps how many plain tags you may combine per query: danbooru/aibooru/e621 = 2, gelbooru/rule34 = 10 (metatags like order:rank or rating:safe do not count). Exceeding the limit returns an error; raise a provider's cap with the BOORU_MAX_TAGS_ env var.

• When unsure how a concept is tagged, call autocomplete_tags first. Use this tool when you want to inspect the original tags before deciding how to process them. Each post carries its raw tag_string, the artist/character/copyright splits (when available), and optionally preview/file URLs.

Search an image booru (Danbooru/AIBooru/e621/Gelbooru/Rule34) by TAG and return ready-to-use, cleaned AI-art prompts. Each result has the full prompt string plus its category split (appearance/clothing/pose/scenery/character/quality/other). This is the recommended tool for finding 'ready-to-use AI painting tags'.

== WHEN TO USE A DIFFERENT TOOL INSTEAD ==

• 'What does character X look like?' / 'list costume variants of X' / 'what tags describe X' -> use danbooru_get_character_profile (one call: trait frequency table + wiki + bidirectional implications). It is built for describing a known character, while search_prompts returns sample images each turned into a prompt — different goals.

• 'Find the canonical tag name for a Chinese/Japanese term or a misspelling' -> use autocomplete_tags (alias/fuzzy/other_names aware).

== SEARCH SEMANTICS — read this before calling ==

• This is NOT a keyword-search engine like Google. You query with one or more booru TAGS, and results are images ALL of whose tags are present.

• A booru tag is a single token, multi-word tags use UNDERSCORE: 'blue_hair', 'hatsune_miku', 'sitting_on_chair' — never 'blue hair'.

• MULTI-TAG SEARCH IS SUPPORTED but each provider caps how many plain tags you may combine per query: danbooru/aibooru/e621 = 2, gelbooru/rule34 = 10 (metatags like order:rank or rating:safe do not count). Exceeding the limit returns an error telling you how to fix it; you can also raise a provider's cap with the BOORU_MAX_TAGS_ env var (e.g. BOORU_MAX_TAGS_DANBOORU=6 for a Gold account).

• STRONGLY PREFER A SINGLE TAG to start. 'hatsune_miku' returns thousands of well-tagged images; 'hatsune_miku blue_hair school_uniform smile' is treated as AND → dramatically fewer results (often zero). The cleaned prompt already contains far more detail than your search tag, so adding search tags is at best redundant and at worst returns nothing.

• When you need a pure-quality / generalized set (no specific character or franchise yet), search '1girl' or '1boy' alone, or use 'random' order.

• Do not write natural-language queries ('a girl with blue hair sitting in a classroom') — that is not supported. Translate to booru tags first.

• Use autocomplete_tags if you are not sure how a concept is spelled as a booru tag (e.g. user said '法国女仆' → try 'french', 'maid' and let autocomplete return 'french maid', 'maid_uniform', etc.).

• Common useful single-tag starting points: a character tag ('hatsune_miku'), an artist tag, a copyright tag ('blue_archive'), a character + single qualifier ('2girls' for group shots), or 'order:rank' / 'random' for non-themed browsing via order param.