> ## Documentation Index > Fetch the complete documentation index at: https://aisa.one/docs/llms.txt > Use this file to discover all available pages before exploring further. # 蒲公英笔记博主广场（找博主 > 蒲公英笔记博主广场（找博主 蒲公英笔记博主广场（找博主 **计费** — 成功调用基准价 \$0.029（= provider 成本 × 1.45；最终扣费按你的套餐倍率浮动）。仅当上游响应体 `code` 为 `200` 时计费。 响应结构尚未完整文档化 —— **以实际调用为准**。 ## 示例 ```bash theme={null} curl -X POST "https://api.aisa.one/apis/v1/tikhub/xiaohongshu/pgy/get_blogger_list" \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{}' # 请求体结构见下方 schema ``` ## OpenAPI ````yaml openapi/tikhub.json POST /tikhub/xiaohongshu/pgy/get_blogger_list openapi: 3.1.0 info: title: AIsa API description: >- Capability layer for the agentic economy. Models, skills, payments, and deployment — everything AI agents need to reason, act, and transact. This spec consolidates all AIsa API endpoints into a single reference. version: 1.0.0 contact: name: AIsa url: https://aisa.one email: developer@aisa.one license: name: MIT url: https://opensource.org/licenses/MIT termsOfService: https://aisa.one/tos servers: - url: https://api.aisa.one/apis/v1 description: AIsa Data APIs (Bearer auth — register at https://aisa.one) - url: https://api.aisa.one/apis/v2 description: >- AIsa Data APIs (x402 pay-per-call) — same surface as /apis/v1, mirrored. No registration; receive HTTP 402 challenge, settle with stablecoin micropayment. Spec: https://www.x402.org. Open-source gateway implementation: https://github.com/AIsa-team/aisa-proxy x-implementation: https://github.com/AIsa-team/aisa-proxy - url: https://api.aisa.one/v1 description: AIsa LLM Inference (OpenAI-compatible, Bearer auth) - url: https://api.aisa.one/v1beta description: AIsa Gemini-compatible GenerateContent (Bearer auth) security: - BearerAuth: [] tags: - name: AI Models description: >- Access 50+ LLMs via OpenAI-compatible, Anthropic, and Google Gemini interfaces - name: Account & Usage description: >- Programmatic account balance and usage endpoints for cost monitoring and alerting - name: Agent Email description: >- AI-agent email accounts, inboxes, threads, drafts, and message send/reply via AgentMail.to - name: Crypto Data description: Cryptocurrency prices, markets, and exchange data via CoinGecko - name: Financial Data description: Stock prices, financials, analyst estimates, SEC filings, and macro data - name: Image Generation description: Generate and edit images using AI models - name: Instagram description: >- Read public Instagram data — profiles, posts, reels, highlights, comments, and Google-backed search - name: Market Intelligence description: '' - name: Other description: '' - name: Pinterest description: >- Read public Pinterest data — search pins, fetch pin details, and browse boards - name: Prediction Markets description: Query prediction markets — Polymarket, Kalshi, and matching markets - name: Reddit description: >- Read public Reddit data — search posts, browse subreddits, and fetch comments - name: SEO & Search Data description: >- SERP, keywords, backlinks, domain and competitor analysis — DataForSEO, Semrush, and Ahrefs - name: Sales Intelligence description: B2B contact and company enrichment, search, and outreach via Apollo.io - name: Scholar Search description: Search academic papers and research - name: Twitter / X description: >- Read, search, and interact with Twitter/X — profiles, tweets, communities, trends, and engagement - name: WaveInflu description: '' - name: Web & News Search description: >- Search the web and news — Tavily search, Oxylabs AI-answer-engine queries, and model-grounded search - name: YouTube Search description: Search YouTube videos paths: /tikhub/xiaohongshu/pgy/get_blogger_list: post: tags: - Other summary: PGY note blogger square description: PGY note blogger square operationId: post_tikhub_xiaohongshu_pgy_get_blogger_list requestBody: content: application/json: schema: $ref: '#/components/schemas/BloggerListReq' required: true responses: '200': description: >- Successful response. Response structure varies; refer to the actual API response. components: schemas: BloggerListReq: properties: page_num: type: integer maximum: 250 minimum: 1 title: Page Num description: Page, 1-250 default: 1 examples: - 1 page_size: type: integer maximum: 20 minimum: 1 title: Page Size description: 'Page size, 1-20 (web fixed at 20; recommended: keep at 20)' default: 20 examples: - 20 keyword: type: string maxLength: 50 title: Keyword description: Keyword; empty=filter only, no search default: '' examples: - Coffee search_type: type: integer enum: - 0 - 1 title: Search Type description: 1=search bloggers by note keyword (default), 0=search by nickname default: 1 examples: - 1 column: type: string enum: - comprehensiverank - fansCount - price - clickNum - mEngagementNum - accumCommonImpMedinNum30d - accumPicCommonImpMedinNum30d - accumVideoCommonImpMedinNum30d - pictureClickMidNum - videoClickMidNum - pictureInterMidNum - videoInterMidNum - thousandLikePercent30 - pictureThousandLikePercent30 - videoThousandLikePercent30 - hundredLikePercent30 - pictureHundredLikePercent30 - videoHundredLikePercent30 - videoFinishRate - accumCoopImpMedinNum30d - readMidCoop30 - interMidCoop30 - mCpuvNum30d - picturePrice - videoPrice - pictureReadCost - videoReadCost - inviteReply48hNumRatio - fans30GrowthRate - fansActiveIn28dLv - fansEngageNum30dLv - kliveCnt30d - avgLiveViewerNum - avgAgmv90d title: Column description: Sort column, 34 options, see endpoint docs default: comprehensiverank examples: - comprehensiverank sort: type: string enum: - desc - asc title: Sort description: 'Sort direction: desc/asc' default: desc examples: - desc brand_user_id: anyOf: - type: string - type: 'null' title: Brand User Id description: Brand/advertiser user ID, optional examples: - 5dcfa5370000000001006030 similar_user_id: anyOf: - type: string - type: 'null' title: Similar User Id description: >- Find similar bloggers; unlike /get_blogger_similar it can be combined with all filters below examples: - 5c668b3e0000000012021605 similar_word: anyOf: - type: string maxLength: 64 - type: 'null' title: Similar Word description: Nickname echo for similar_user_id; does not affect results goal: anyOf: - $ref: '#/components/schemas/GoalFilter' - type: 'null' description: Cooperation goal filters blogger: anyOf: - $ref: '#/components/schemas/BloggerAttrFilter' - type: 'null' description: Blogger attribute filters fans: anyOf: - $ref: '#/components/schemas/FansFilter' - type: 'null' description: Fans profile filters note: anyOf: - $ref: '#/components/schemas/NoteFilter' - type: 'null' description: Organic note metric filters coop: anyOf: - $ref: '#/components/schemas/CoopFilter' - type: 'null' description: Sponsored note metric filters live: anyOf: - $ref: >- #/components/schemas/crawlers__xiaohongshu__pgy__pgy_filters__LiveFilter - type: 'null' description: Live metric filters flags: anyOf: - $ref: >- #/components/schemas/crawlers__xiaohongshu__pgy__pgy_filters__FlagFilter - type: 'null' description: Boolean flag filters additionalProperties: false type: object title: BloggerListReq GoalFilter: properties: market_target: anyOf: - type: string enum: - estimateAllCpm - mAccumImpNum - readCost - mValidRawReadFeedNum - estimateEngageCost - mEngagementNum90d - estimateCpuv - mCpuvNum - type: 'null' title: Market Target description: >- Marketing objective: on the web UI this is a three-level selector (objective -> metric -> measurement method); pass the final leaf code: `estimateAllCpm` Awareness/Impression Performance/Cost, `mAccumImpNum` Awareness/Impression Performance/Scale, `readCost` Awareness/Read Performance/Cost, `mValidRawReadFeedNum` Awareness/Read Performance/Scale, `estimateEngageCost` Consideration/Engagement Performance/Cost, `mEngagementNum90d` Consideration/Engagement Performance/Scale, `estimateCpuv` Conversion/Overflow Store-Visit Performance/Cost, `mCpuvNum` Conversion/Overflow Store-Visit Performance/Scale examples: - mAccumImpNum additionalProperties: false type: object title: GoalFilter description: >- Cooperation goal: match bloggers by marketing objective (Exposure / Seeding / Conversion). To filter bloggers by audience segment, use `BloggerAttrFilter.audience20` (top-20 audiences). BloggerAttrFilter: properties: content_tag: anyOf: - items: type: string maxLength: 64 type: array maxItems: 30 - type: 'null' title: Content Tag description: >- Creator category: use the category name. There are 28 level-1 categories: Beauty / Skincare / Personal Care / Parenting / Fashion / Food / Home Decor / Film & TV & Entertainment News / Sports & Fitness / Pets / Culture & Art / Hobbies / Daily Life / Education / Workplace / Relationships / Photography / Gaming / Tech & Digital / Travel / Music / Comedy / Health & Wellness / Automotive / Wedding / Business & Finance / Stock Footage / Other. Level-2 names may also be used; **both levels can be mixed in the same array**, and multiple values are OR-ed. To select an entire category, use its level-1 name — **do not expand it to all level-2 names** (some creators are tagged at level-1 only; expanding will miss them). **If all categories are acceptable, omit this field entirely** — there is no category named "All", and passing one will silently return 0 results. Incorrect names also return 0 results without an error; always copy names verbatim from the contentTags dictionary. Note: this is a separate classification from "industry" (coop.industry). examples: - - Beauty - Facial Care - Facial Cleanser personal_tags: anyOf: - items: type: string maxLength: 64 type: array maxItems: 30 - type: 'null' title: Personal Tags description: >- Creator persona: use the `tagName` (bare name) from personalTagsV2 in `/get_blogger_filter_options`. This is a three-level tree (family identity / professional identity / background -> group -> leaf). **Any level can be filled in, and levels can be mixed**; multiple values are OR-ed. To select an entire group, use the group name — no need to enumerate every leaf (e.g. level-1 `Professional Identity`=91 results, level-2 `Internet`=8, leaf `Mom`=31). examples: - - Mother feature_tags: anyOf: - items: type: string maxLength: 64 type: array maxItems: 30 - type: 'null' title: Feature Tags description: >- Specialty content: use the bare name from featureTags in `/get_blogger_filter_options`. **Level-1 (format / style / lifestyle / skin type & tone / skincare) and level-2 (e.g. vlog / Korean style) can both be used and mixed**; multiple values are OR-ed. Selecting a level-1 name includes the entire group (e.g. level-1 `Format`=183 results, level-2 `vlog`=136). examples: - - vlog content_theme: anyOf: - items: type: string maxLength: 64 type: array maxItems: 30 - type: 'null' title: Content Theme description: >- Content theme: use the **level-1 label** from contentTheme in `/blogger/filter-options` (e.g. Automotive / General). Note: only level-1 is supported; using a sub-item label or ID returns no results. examples: - - Automotive Features audience20: anyOf: - items: type: string maxLength: 64 type: array maxItems: 30 - type: 'null' title: Audience20 description: >- "Top-20 Audience Segments": **do not use bare flat names**. For level-1, provide just the name (e.g. `Outdoor Leisure`, = the entire segment); **for level-2, write `"level1 level1-level2"`** (e.g. `"Outdoor Leisure Outdoor Leisure-Extreme Challengers"`) — the level-1 name must appear twice. **Using the level-2 label directly from the dictionary (e.g. `Extreme Challengers`) silently returns 0 results**. Level-1 and composite level-2 values can be mixed in the same array; multiple values are OR-ed. Full list available from `/get_blogger_filter_options` under audience20 (19 level-1 / 121 level-2 segments). examples: - - Free Travel - Outdoor Freedom - Extreme Challengers industry_crowd: anyOf: - items: type: string maxLength: 64 type: array maxItems: 30 - type: 'null' title: Industry Crowd description: >- Industry-specific audience profile: use the **deepest leaf value** from automotiveIndustryTag in `/blogger/filter-options` (numeric string ID, e.g. 15795914; 1,600+ available). Using a label or intermediate-level ID returns no results. examples: - - '15795914' consumer_behavior: anyOf: - items: type: string maxLength: 64 type: array maxItems: 30 - type: 'null' title: Consumer Behavior description: >- Estimated consumer behavior: **must use the full space-separated path including the root node** (e.g. `"Content Behavior Estimate Automotive"`); the path can stop at any depth. See **consumerBehavior** in `/get_blogger_filter_options` (5-level tree). WARNING: Using a bare leaf name causes no error and returns no 0 results — instead the filter is silently ignored and the full unfiltered result set is returned. Only the full path form actually filters (e.g. `["Content Behavior Estimate Automotive"]` returns 54 results). When using this filter, compare totals with and without it to confirm the filter is taking effect. Note: this field is rolled out per account; if the dictionary returns null, the current account does not have access. examples: - - 'Content behavior estimate: Automotive' note_scene: anyOf: - items: type: string maxLength: 64 type: array maxItems: 30 - type: 'null' title: Note Scene description: >- Note category: use the **full space-separated path** (Category Level1 Level2 Level3), e.g. "Automotive Rational-Decision Car-Selection-Guide Policy". The path can stop at any depth (deeper = narrower). **Using only the last segment (e.g. "Car-Selection-Guide") returns 0 results.** See **contentScenes** in `/get_blogger_filter_options` (e.g. `["Automotive"]`=75, `["Automotive Rational-Decision"]`=23, 4-level path=1, `["Car-Selection-Guide"]`=0). examples: - - >- Automotive: Rational Decision-Making, Car Selection Guide, Policy gender: anyOf: - type: string enum: - Male - Female - type: 'null' title: Gender description: 'Gender: Male / Female' examples: - Female location: anyOf: - items: type: string maxLength: 64 type: array maxItems: 30 - type: 'null' title: Location description: >- Creator location: plain place names work at any level (province/city/district), and multiple values are OR-ed (e.g. ["Shanghai"], ["Hangzhou"], ["Shangcheng District"]). Full path format (e.g. "China Zhejiang Hangzhou") is also accepted and will be converted automatically. See **areas** in /get_blogger_filter_options for available place names (down to district level). examples: - - Shanghai signed: anyOf: - type: integer enum: - -1 - 0 - 1 - type: 'null' title: Signed description: >- Contract status: -1=no filter, 0=individual creator, 1=agency-signed creator examples: - 0 additionalProperties: false type: object title: BloggerAttrFilter description: >- Blogger attributes: category, persona, content specialty, audience profile, gender, region, contract status, etc. For hierarchical options (category / persona / content specialty / content theme / top-20 audiences / industry profile / estimated consumer behavior), first call `/api/v2/pgy/blogger/filter-options` to get the values — each field description specifies which level to use. FansFilter: properties: count: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Follower count range age: anyOf: - type: integer enum: - 0 - 1 - 2 - 3 - 4 - 5 - type: 'null' title: Age description: >- Audience age distribution: 0=no filter, 1=under 18, 2=18-24, 3=25-34, 4=35-44, 5=over 44 (high proportion) examples: - 3 gender: anyOf: - type: integer enum: - 0 - 1 - 2 - type: 'null' title: Gender description: >- Audience gender: 0=no filter, 1=predominantly male, 2=predominantly female examples: - 2 marital_status: anyOf: - type: integer enum: - -1 - 0 - 1 - type: 'null' title: Marital Status description: 'Relationship status: -1=no filter, 0=single/unmarried, 1=married' consumption_level: anyOf: - type: integer enum: - -1 - 0 - 1 - 2 - type: 'null' title: Consumption Level description: 'Consumption level: -1=no filter, 0=low, 1=medium, 2=high' child_age: anyOf: - items: type: integer enum: - 0 - 2 - 3 - 4 - 5 - 6 - 7 - 8 type: array maxItems: 30 - type: 'null' title: Child Age description: >- Parenting stage: 0=trying to conceive, 2=0-6 months, 3=7-12 months, 4=1-3 years, 5=4-6 years, 6=7-12 years, 7=early pregnancy, 8=late pregnancy device_price: anyOf: - items: type: string enum: - 0-999 - 1000-1999 - 2000-2999 - 3000-3999 - 4000-4999 - 5000-5999 - 6000-6999 - 7000-7999 - 8000+ type: array maxItems: 30 - type: 'null' title: Device Price description: Phone price tier, e.g. 3000-3999 / 8000+ examples: - - 3000-3999 device_brand: anyOf: - items: type: string enum: - Apple - Huawei - OPPO - VIVO - Honor - Xiaomi - OnePlus - Meizu - ZTE - Autocomplete type: array maxItems: 30 - type: 'null' title: Device Brand description: Phone brand examples: - - Apple location: anyOf: - items: type: string maxLength: 64 type: array maxItems: 30 - type: 'null' title: Location description: >- Audience location: plain place names work (e.g. ["Shanghai"], ["Hangzhou"]) and the full path will be filled in automatically. Full path format (e.g. "China Zhejiang Hangzhou") is also accepted. Note: this tree only goes down to city level (no district). See **fansAreas** in /get_blogger_filter_options for available values. examples: - - Shanghai additionalProperties: false type: object title: FansFilter description: >- Audience profile — the "Audience Profile" section of the web page (follower count / age / gender / region / relationship status / spending / parenting & baby / mobile device). NoteFilter: properties: imp_median: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Median impressions read_median: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Median reads inter_median: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Median engagement count thousand_like_percent: anyOf: - $ref: '#/components/schemas/PercentRange' - type: 'null' description: >- 1,000-like note ratio: fill 0-100 percent points (20 means 20%, not 0.2) note_type: anyOf: - type: integer enum: - 0 - 1 - 2 - type: 'null' title: Note Type description: >- Note type: 0=no filter, 1=primarily image+text notes, 2=primarily video notes examples: - 2 additionalProperties: false type: object title: NoteFilter description: >- Organic note data — the 'Daily Notes' section on the web page (organic traffic performance over the last 30 days). CoopFilter: properties: pic_price: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Cooperation quote — fixed price for image+text notes (RMB) video_price: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Cooperation quote — fixed price for video notes (RMB) order_count: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: >- Open orders (accepted but not yet completed), NOT lifetime collaboration count. This reflects current in-progress orders; do not use this field to find experienced creators. reply_48h_ratio: anyOf: - $ref: '#/components/schemas/PercentRange' - type: 'null' description: >- Cooperation credit — 48h invitation reply rate: fill 0-100 percent points (80 means 80%) industry: anyOf: - type: string enum: - No filter - Beauty & Personal Care - Food & Beverages - Parenting & Baby - Consumer Electronics & Appliances - Daily necessities - Clothing and accessories - Internet - Lifestyle Services - Home & Building Materials - Automotive - type: 'null' title: Industry description: Recent collaboration industries examples: - Beauty & Personal Care brand_ids: anyOf: - items: type: string maxLength: 64 type: array maxItems: 20 - type: 'null' title: Brand Ids description: 'Recent collaboration brands: list of brand userIds' exclude_brands: anyOf: - type: boolean - type: 'null' title: Exclude Brands description: >- Used with brand_ids: true=exclude bloggers who have cooperated with these brands imp_median: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Reach scale — median impressions (collaboration notes) read_median: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Reach scale — median reads (collaboration notes) inter_median: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Reach scale — median engagement count (collaboration notes) cpuv: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: >- Reach scale — median store-visit UV count (not cost): median store-visit UV count for cross-domain collaboration notes in the last 30 days. **This is a visit count, not a monetary amount** (web tiers: 500-1000 / 1000-2000 / 2000-5000 / 5000-10000 / 10000+). To filter by cost per store visit, use estimate_cpuv instead. pic_cpm: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Estimated CPM — image+text notes video_cpm: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Estimated CPM — video notes pic_read_price: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Estimated cost per read — image+text notes video_read_price: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Estimated cost per read — video notes pic_engage_cost: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Estimated cost per engagement — image+text notes video_engage_cost: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Estimated cost per engagement — video notes estimate_cpuv: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Overflow store-visit cost per visit additionalProperties: false type: object title: CoopFilter description: >- Sponsored notes — the "Sponsored Notes" section on the web page (cooperation performance + data performance). crawlers__xiaohongshu__pgy__pgy_filters__LiveFilter: properties: live_count: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: Number of live streams in the last 30 days avg_viewer: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: >- Average viewers per live stream over the last 90 days (live_count uses a 30-day window instead). avg_gmv: anyOf: - $ref: '#/components/schemas/RangeFilter' - type: 'null' description: >- Average GMV per live stream over the last 90 days (live_count uses a 30-day window instead). additionalProperties: false type: object title: LiveFilter description: Live data — the "Live Data" section of the web page. crawlers__xiaohongshu__pgy__pgy_filters__FlagFilter: properties: is_star: anyOf: - type: boolean - type: 'null' title: Is Star description: Celebrity new_high_quality: anyOf: - type: boolean - type: 'null' title: New High Quality description: Rising creator high_quality: anyOf: - type: boolean - type: 'null' title: High Quality description: Quality creator note_and_live: anyOf: - type: boolean - type: 'null' title: Note And Live description: Both notes and live streams available for collaboration filter_intention: anyOf: - type: boolean - type: 'null' title: Filter Intention description: Intended industry match activity_codes: anyOf: - items: type: string maxLength: 64 type: array maxItems: 30 - type: 'null' title: Activity Codes description: >- Trending activity: use the `code` value for each item in the activities list from `/blogger/filter-options` exclude_low_active: anyOf: - type: boolean - type: 'null' title: Exclude Low Active description: Exclude low-activity creators exclude_fans_down: anyOf: - type: boolean - type: 'null' title: Exclude Fans Down description: Exclude creators with declining followers exclude_cooperated: anyOf: - type: boolean - type: 'null' title: Exclude Cooperated description: Exclude previously collaborated creators exclude_invited: anyOf: - type: boolean - type: 'null' title: Exclude Invited description: Exclude already-invited creators industry_first: anyOf: - type: string maxLength: 64 - type: 'null' title: Industry First description: >- Industry-recommended creators — level-1 industry name (e.g. Beauty & Personal Care); the platform will recommend creators specializing in that industry examples: - Beauty & Personal Care industry_second: anyOf: - type: string maxLength: 64 - type: 'null' title: Industry Second description: >- Industry-recommended creators — level-2 industry name (used together with industry_first; values from taxonomy2Tags in industryTags in `/blogger/filter-options`) additionalProperties: false type: object title: FlagFilter description: >- Boolean flags — controls the "Platform Recommendations" (featured bloggers / trending campaigns) and "Standard Exclusions" (one-click exclude) sections on the web page. RangeFilter: properties: min: anyOf: - type: number maximum: 1000000000000 minimum: 0 - type: 'null' title: Min description: Minimum value (inclusive) examples: - 10000 max: anyOf: - type: number maximum: 1000000000000 minimum: 0 - type: 'null' title: Max description: Maximum value (inclusive); omit for no upper limit examples: - 50000 additionalProperties: false type: object title: RangeFilter description: >- Numeric range corresponding to the custom input box in the web filter. - Either bound may be omitted: omitting `min` defaults to 0, omitting `max` means no upper limit - At least one bound must be provided, otherwise 422 PercentRange: properties: min: anyOf: - type: number maximum: 100 minimum: 0 - type: 'null' title: Min description: Minimum value (inclusive), percentile 0-100 examples: - 20 max: anyOf: - type: number maximum: 100 minimum: 0 - type: 'null' title: Max description: Maximum value (inclusive), percentile 0-100 examples: - 80 additionalProperties: false type: object title: PercentRange description: >- Percentage range: enter a value between 0 and 100 as a percentile, e.g. `40` means 40% (**not 0.4**). This convention follows the PGY web interface: the tier for notes-per-thousand-likes is written as `{"40% and above": [40, null]}` or `{"below 10%": [null, 10]}` — the value sent upstream is simply 40 or 10. **What is and is not validated — stated clearly to avoid confusion**: * Caught: values with the wrong order of magnitude — entering 200 or 9500 (as if using parts-per-ten-thousand or raw counts) will return a 422 error, whereas the generic RangeFilter allows up to 1e12 and previously passed such values through. * **Not caught: values that are too small** — entering 0.4 is still valid because 0.4% is a legitimate value; the type system cannot distinguish "meant 40% but wrote 0.4" from "genuinely filtering for 0.4%". This ambiguity can only be resolved through field documentation. securitySchemes: BearerAuth: type: http scheme: bearer description: AIsa API key. Get yours at https://aisa.one ````