AIKIZI
Ramakrishna Sai Sharan v1.2.0
Find and decode image styles
From the developer
Declared capabilities
- Search published style decodes
- Read a style breakdown and its prompt
- Decode your own image from a link or a picture you drop in
Cost & access
AIKIZI sells nothing inside ChatGPT. Decoding uses credits the person already has on their AIKIZI account; the plugin has no prices, checkout or purchase links.
Declared limitations
- No purchases: AIKIZI has no tool to buy credits or change a plan, and sells nothing inside ChatGPT.
- No image generation: the plugin finds, reads and decodes styles; it does not make new images.
- No access to another person's library: the tools read only published decodes and the connected account's own decodes.
Onboarding example
Find moody neon night city styles and show the best one's breakdown
Sources & dates
Developer declarations; not independently tested. Listing last checked Nov 14, 2026. Package observed Nov 14, 2026.
Description
From the marketplace listing
AIKIZI is a library of decoded image styles. Search thousands of published style breakdowns by look, subject, mood or colour, and open any one to see its styles, style DNA, moods, key tokens, colour hues and a generation prompt you can start from. Searching and reading published decodes needs no account. Connect your AIKIZI account to decode your own image: drop, pick or paste a picture into the drop box that opens in the chat, or give a public link. You can also check your credit balance and read the decodes in your library. Decoding uses credits from your AIKIZI account, and every decode you make stays private to you. For designers, art directors, illustrators and anyone writing image prompts who wants to understand how a picture was made. AIKIZI does not generate images here, never publishes or deletes anything, and only reads decodes that are published or that belong to the connected account.
Language: English · Automatically detected from descriptions.
Matches for “code generation”
Exact text from the indicated source. A mention alone does not establish support for your task.
Publisher full description
AIKIZI is a library of decoded image styles. Search thousands of published style breakdowns by look, subject, mood or colour, and open any one to see its styles, style DNA, moods, key tokens, colour hues and a generation prompt you can start from. Searching and reading published decodes needs no account. Connect your AIKIZI account to decode your own image: drop, pick or paste a picture into the drop box that opens in the chat, or give a public link. You can also check your credit balance and read the decodes in your library. Decoding uses credits from your AIKIZI account, and every decode you make stays private to you. For designers, art directors, illustrators and anyone writing image prompts who wants to understand how a picture was made. AIKIZI does not generate images here, never publishes or deletes anything, and only reads decodes that are published or that belong to the connected account.
Publisher subtitle
Find and decode image styles
Publisher capabilities · listing
Search published style decodes Read a style breakdown and its prompt Decode your own image from a link or a picture you drop in Read your AIKIZI library and balance
Publisher keywords · listing
aikizi image style decode prompt palette art direction
Publisher description
Search decoded image styles and their prompts on aikizi.com, break down any published decode, and with your own AIKIZI account decode an image you drop into the chat or give as a link, and read your library.
Changes
No changes recorded yet. Follow this plugin to keep track of future changes.
First seen Nov 12, 2026. Last seen Nov 14, 2026.
Similar plugins
Find competitors to follow as you develop your plugin.
-
Find a public PromptKosh style or use a style ID from promptkosh.in. PromptKosh sends the style prompt and any required inputs to ChatGPT...
Shared terms: style, image, prompt, find
-
Make and edit pictures and video with Plnty's tools, from the chat.
Shared terms: image, here, library, made
-
Generate and edit images and video with Batch Banana from your agent, and run batches of many prompts. Pick a model and a style, generate...
Shared terms: image, library, pick, prompt
-
Search LivePair's prompt library — tested AI image and video prompts with the results they produced.
Shared terms: image, library, prompt, style
-
Turn an image from your MegaScenes library into a short video clip: describe the motion and the camera move.
Shared terms: image, here, library, made
Keywords
Files & skills
File archives
Skill instructions
aikizi-decode2.76 KB
--- name: aikizi-decode description: Decode the person's own image on AIKIZI with their account, read their own decodes and library, or check their credits. Use when they ask to decode a picture they have or an image link, see what they made on AIKIZI, or ask about their balance. These tools need an AIKIZI sign-in; decode_image uses the account's credits. --- # AIKIZI with the person's account Eight tools act on the signed-in AIKIZI account. The person connects once through the AIKIZI sign-in; nothing is shared until they allow it. Decoding uses credits from that account, so be deliberate. AIKIZI does not generate images here; when the person asks for a new image, give them the decode's prompt to use. ## Before spending - Run `check_balance` first when the person has not decoded in this conversation. Tell them the balance in one line. If it is too low for a decode, stop and say so; do not retry. - Say what you are about to decode and why, in one sentence, before the first decode. After that, continue without re-asking unless they change the task. ## Decode an image 1. A picture the person has (attached to the chat or on their device): call `add_image`. A drop box opens in the chat; they drop, pick or paste the image (up to 10 MB; SVG, GIF and other formats become PNG) and the box decodes it there and shows the result. Attachments cannot reach AIKIZI directly, so never ask them to host the file somewhere. Adding spends nothing; the decode uses credits. 2. A public `https://` link to a JPEG, PNG or WebP up to 10 MB: `decode_image` with `image_url`. Data URLs, local paths and private links fail. `decode_image` also takes the `image_id` that `add_image` returned. 3. If the answer says "still running", call `check_job_status` with the `decode_id` every 10 seconds or so, for up to two minutes. Then read the result with `get_my_decode`. 4. Present styles, style DNA, mood, light, colours, key tokens and the prompt. Decodes stay private in their library. ## Their library - `list_my_creations` with `type` image, video or audio and a small `limit` for "what did I make". - `get_my_decode` for one decode (published or private); `get_creation` for one image they made on aikizi.com. Show the decode's own `preview_image_url`, never a similar image from the web. - `place_in_brain` with a `decode_id` lists the public AIKIZI Brain pictures closest to one of their decodes. It only reads: it adds nothing to the Brain, publishes nothing and spends nothing; their decode stays private. ## Never - Never call `decode_image` to test the connection; `check_balance` is the test. - Never claim a decode or creation is published. Everything a decode makes stays private. - Never show another person's private decodes; only published decodes and this account's own are readable.
aikizi-style-reference2.1 KB
---
name: aikizi-style-reference
description: Find and explain image styles with AIKIZI. Use when the person wants a visual reference, a style breakdown, a colour palette, a mood, "how was this made", pictures that look alike, or a prompt to start an image from. Uses the public search_styles, get_style_decode and find_look_alikes tools; no sign-in needed.
---
# AIKIZI style reference
AIKIZI (aikizi.com) decodes images the way an art director reads them: styles, style DNA, subjects, moods,
key tokens, colour hues and a generation prompt. The three public tools read only decodes their owners published.
## Flow
1. `search_styles` with a short query made of style, subject, mood or colour words (2 to 6 words works best).
Ask for 5 to 8 results. Example: `{"query": "neon noir portrait", "limit": 8}`.
2. Pick the two or three results that fit the ask best. Call `get_style_decode` with each `id` for the full
breakdown. An `aikizi.com/decode/<id>` link the person pastes also works as the id.
3. Answer with what the decode says, in this order: styles and style DNA, mood, key tokens, colour hues, then the
generation prompt as a block the person can copy.
4. Always give the `aikizi.com/decode/<id>` link for each decode you used, so the person can see the image.
5. When you show a decode's picture, show its own `preview_image_url` from AIKIZI. Never show a similar image found
on the web as if it were the decode.
6. For pictures like one decode, call `find_look_alikes` with its `id`: up to 21 published decodes from the AIKIZI
Brain, closest first.
## Rules
- Quote the decode's prompt as given; mark your own edits as yours.
- A search with no good match: try one simpler query (drop the rarest word), then say so. Do not invent styles.
- Decodes belong to the people who published them. Describe the style; do not present the image as the person's own.
- These tools never need the person's account. If they want to decode their own image, use the `aikizi-decode`
skill, which needs an AIKIZI sign-in.
- AIKIZI does not generate images here. When the person wants a new image, give them the decode's prompt to use.
Release notes
1.2.0: place_in_brain is now titled "Find Brain pictures near my decode", and its description says only what it does: it lists the public AIKIZI Brain pictures closest to one of your own decodes and adds, publishes, decodes and spends nothing. get_style_decode, get_my_decode and decode_image now give the decode's own AIKIZI picture in their text answer, and the tool descriptions ask the assistant to show that picture rather than a similar image from the web. find_look_alikes no longer says the Brain holds every public decode (it is the published decodes, rebuilt nightly). The skills name the AIKIZI Brain tools and the preview rule. New: add_image opens a drop box in the chat, so a person can decode a picture from their device (drop, pick or paste it); decode_image also takes the image_id it returns. The in-chat views have a quieter look; the decode card links its picture to its aikizi.com page and has More like this, which shows the AIKIZI Brain look-alikes of that decode as a picture wall. Same server URL, sign-in and review account as 1.1.0; eleven tools (add_image is new).
Declared in the saved package. Remote tools may change independently.
Package details
Publisher declarations from the archived package. These are separate from our research and the live service's terms.
- Package license
- MIT
- Package author
- Ramakrishna Sai Sharan
- Keywords
- See keywords
- Commerce declaration
- Does not support commerceThis does not establish whether access is free or paid.
- Review scenarios
- 8 positive · 3 negativeScenario counts, not test results.
Declared capabilities
- Search published style decodes
- Read a style breakdown and its prompt
- Decode your own image from a link or a picture you drop in
- Read your AIKIZI library and balance
Package observed Nov 14, 2026.
Technical details
- First seen
- Nov 12, 2026 · 18:00 UTC
- Last seen
- Nov 14, 2026 · 18:00 UTC
- Collection status
- Collected
plugin_asdk_app_6aac0a15e9f881918b3493fec13e773e
Download plugin data (JSON)Plugin details: served from cache. Prepared . Refreshed when plugin data changes or after 6 hours. Your account and Follow status are live.