ImgKits MCP server
io.github.yichong1/mcp-imgkitsyichong1streamable-httpv1.0.1Scanned
ImgKits MCP: translate, blind watermark, audio, PDF→MD→translate, text utils, colorize.
21 tools: 4 do something that can't be undone, 15 change things and 2 only read.
- Reads private data · 0
- Reads text others write · 18
- Can send data out · 4
- 2 findings
Tools listed without sign-in. Remote posture
With Anlyon
4 tools here do something that can't be undone. Govern them.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the ImgKits MCP server can do
The ImgKits MCP server is published by the GitHub account yichong1. It has 21 tools. It can do things that cannot be undone, such as sending, paying or publishing (translate_text, translate_document, translate_image and translate_subtitle) and create or change data (15 tools: convert_to_markdown, ocr_image, extract_tables, detect_language and 11 more), but cannot delete data. 2 tools only read.
18 tools bring in text other people can write (convert_to_markdown, ocr_image, extract_tables and 15 more); 4 tools can send data out of your system (translate_text, translate_document, translate_image and 1 more).
The fixed checks matched 2 findings (annotation-conflict and tool-shadowing). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.
Last scanned Nov 19, 2026, version 1.0.1.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
21 tools
Can't be undone 4Claude Code: ask
translate_documentTranslate a document from a public URL.Reads text others writeCan send data outNo findings
Why it is in this group
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it Translate a document from a public URL. Supported formats include PDF, DOCX/DOC, XLSX/XLS, PPTX/PPT, EPUB, TXT, HTML, MD and CSV. Upstream size limits: PDF <= 30 MB, other formats <= 80 MB. This is an async job: the call waits up to wait_seconds (default 30, max 120) and returns a job_id if still running. Call again with the same job_id to poll; polling does not start a new translation job. Upstream deletes temporary/original files after processing and result URLs expire (about 24 hours), so download promptly. Cost: 5 credits per new job, charged by mcp-credit-gateway before forwarding the request.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
translate_imageTranslate text inside 1-10 images supplied as public URLs (JPG/PNG/WebP recommended; keep each image <= 10 MB).Reads text others writeCan send data outNo findings
Why it is in this group
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it Translate text inside 1-10 images supplied as public URLs (JPG/PNG/WebP recommended; keep each image <= 10 MB). This is an async job: the call waits up to wait_seconds (default 30, max 120) and returns a job_id if still running. Call again with the same job_id to poll; polling does not start a new translation job. Upstream result URLs expire (about 24 hours), so download promptly. Cost: 2 credits per new job, charged by mcp-credit-gateway before forwarding the request.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
translate_subtitleTranslate SRT, WebVTT or ASS/SSA subtitle content (up to 1,000,000 characters).Reads text others writeCan send data outNo findings
Why it is in this group
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it Translate SRT, WebVTT or ASS/SSA subtitle content (up to 1,000,000 characters). Timestamps, cue order, SRT/VTT indices and ASS event fields are preserved exactly; only cue text is translated. Synchronous. Cost: 3 credits per call, charged by mcp-credit-gateway before forwarding the request.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
translate_textTranslate up to 50,000 characters of plain text (up to 50 strings).Reads text others writeCan send data outNo findings
Why it is in this group
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it Translate up to 50,000 characters of plain text (up to 50 strings). Long inputs are split into <=1,900-character chunks automatically. Synchronous. Cost: 1 credit per call; mcp-credit-gateway charges before forwarding the request. Requires Authorization: Bearer sk_live_... on the MCP endpoint.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
Changes things 15Claude Code: ask
clone_voice_ttsClone a voice from `clone_audio_url` and speak `text` via Qwen-TTS.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Clone a voice from `clone_audio_url` and speak `text` via Qwen-TTS. Requires `authorization_confirmed=true` (own voice or written authorization). Pass `text` + `clone_audio_url` (+ recommended `clone_audio_text`) to start, or `job_id` to poll. Cost: 3 credits per call. VOICE AUTHORIZATION: Voice cloning is allowed ONLY for your own voice or a voice you have explicit written authorization to clone. Unauthorized cloning of another person's voice is prohibited. Calls are rate-limited/metered. (声音克隆仅限本人音色或已获书面授权的音色;禁止未授权克隆他人声音;调用限频。) Also: Use only on material you own or are explicitly licensed/authorized to generate or remix. Do not generate or redistribute copyrighted music, voice likenesses, or trademarked audio without permission. Output URLs have a limited TTL and must not be abused for bulk scraping or redistribution. Calls are rate-limited/metered by mcp-credit-gateway and require a valid sk_live_ key. (仅处理自有或已获授权素材;禁止滥用生成/二创他人受版权保护的音乐或声音;输出有 TTL;调用限频。)
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
colorize_photoColorize a photo from a public URL.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Colorize a photo from a public URL. POST {img} to ai.imgkits.com/api/deoldify (AiController Shangse). Black-and-white photos work best; a response with a null URL is treated as failure. Synchronous. Cost: 2 credits per call; mcp-credit-gateway charges before forwarding. Requires Authorization: Bearer sk_live_... on tools/call.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
convert_text_styleTransform text with a pure in-Worker algorithm (no upstream HTTP).1 finding
Why it is in this group
- The name "convert_text_style" says "convert": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "convert_text_style" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "convert_text_style" says "convert": a word for a change.
Description, as the server wrote it Transform text with a pure in-Worker algorithm (no upstream HTTP). Pass required `text` (≤8000 chars) and `mode` one of: morse, binary, old_english, fancy, upper, lower, title, reverse. Modes: morse / binary (mappable chars only; others kept + optional skipped note), old_english (Fraktur Unicode), fancy (Math Sans-Serif Bold Unicode), upper / lower / title / reverse. Returns { text, mode, result }. Cost: 1 credit per call, charged by mcp-credit-gateway before forwarding. Privacy: Text is processed in-memory for this request only and is not stored. Do not send secrets, passwords, or personal data you would not paste into a public tool. Calls are rate-limited/metered by mcp-credit-gateway and require a valid sk_live_ key. (文本仅在本请求内存中处理、不落盘;请勿提交密钥/密码等敏感信息;调用限频。)
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
convert_to_markdownConvert PDF and image files to clean Markdown for RAG, LLM ingestion, and knowledge bases.Reads text others writeNo findings
Why it is in this group
- The name "convert_to_markdown" says "convert": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Convert PDF and image files to clean Markdown for RAG, LLM ingestion, and knowledge bases. Optionally include MinerU layout JSON (content list with blocks and bounding boxes). Input: 1-5 public https file URLs, each ≤20 MiB. Supported: pdf, png, jpeg, webp, bmp, tiff. DOCX is not supported in this release. Private, local, and metadata hosts are rejected. Processing is asynchronous upstream; this call polls until done or times out after about 75 seconds. This Worker does not store uploaded files or results. Upstream MinerU task data is temporary (≤24h). Billing: charged before the MinerU job is created (default 1 credit per tool call). No automatic refunds. Powered by MinerU (opendatalab/MinerU). License: https://github.com/opendatalab/MinerU/blob/master/LICENSE.md
Annotations the server declares
- title: Convert PDF or image to Markdown
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
detect_languageDetect the language of up to 20,000 characters.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Detect the language of up to 20,000 characters. Uses a local Unicode-script + stopword/diacritic heuristic; no upstream call. Returns a language code, confidence, reliability flag and alternatives. Billing: free by default (COST_DETECT_LANGUAGE=0).
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
embed_blind_watermarkEmbed an invisible/blind watermark into an image supplied as a public URL (JPG/PNG/WebP/BMP, <= 10 MB; at least 64x64 px recommended).Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Embed an invisible/blind watermark into an image supplied as a public URL (JPG/PNG/WebP/BMP, <= 10 MB; at least 64x64 px recommended). Uses the AiController EncodeBlindWatermark / Alibaba Cloud IMM path. Pass `watermark_text` (<= 512 chars, e.g. an order id), or omit it to use the server-held BLIND_WATERMARK_KEY secret; the embedded content is never echoed back. `model` is DWT (default, more robust) or DCT; optional `strength` 1-10. Synchronous. Cost: 2 credits per call, charged by mcp-credit-gateway before forwarding.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
extract_tablesExtract tables from PDFs and images as HTML plus a row/column matrix.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Extract tables from PDFs and images as HTML plus a row/column matrix. Merged cells are expanded; only the top-left cell carries content and other covered cells are empty strings. Returns row_count, column_count, page_idx, and bbox for each table. Input: 1-5 public https file URLs, each ≤20 MiB. Supported: pdf, png, jpeg, webp, bmp, tiff. DOCX is not supported in this release. Private, local, and metadata hosts are rejected. Table structure is model-recognized, so complex, borderless, or handwritten tables may contain errors. Processing is asynchronous and this call polls until done or times out after about 75 seconds. This Worker does not store uploaded files or results. Upstream MinerU task data is temporary (≤24h). Billing: charged before the MinerU job is created (default 1 credit per tool call). No automatic refunds. Powered by MinerU (opendatalab/MinerU). License: https://github.com/opendatalab/MinerU/blob/master/LICENSE.md
Annotations the server declares
- title: Extract tables from PDF or image
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
generate_subtitlesTranscribe a public audio URL with Faster-Whisper and assemble SRT or VTT in the Worker.Reads text others writeNo findings
Why it is in this group
- The name "generate_subtitles" says "generate": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Transcribe a public audio URL with Faster-Whisper and assemble SRT or VTT in the Worker. Pass `audio_url` to start or `job_id` to poll; `format` is `srt` (default) or `vtt`. Cost: 2 credits per call. Compliance / SSRF: Use only on material you own or are explicitly licensed/authorized to generate or remix. Do not generate or redistribute copyrighted music, voice likenesses, or trademarked audio without permission. Output URLs have a limited TTL and must not be abused for bulk scraping or redistribution. Calls are rate-limited/metered by mcp-credit-gateway and require a valid sk_live_ key. (仅处理自有或已获授权素材;禁止滥用生成/二创他人受版权保护的音乐或声音;输出有 TTL;调用限频。)
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
ocr_imageExtract text from images, screenshots, and scanned PDF pages.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Extract text from images, screenshots, and scanned PDF pages. Returns text blocks with block-level bounding boxes [x0, y0, x1, y1] and page indexes. Coordinates are MinerU content-list coordinates, not character-level boxes and not rescaled to pixels. Input: 1-5 public https file URLs, each ≤20 MiB. Supported: pdf, png, jpeg, webp, bmp, tiff. DOCX is not supported in this release. Private, local, and metadata hosts are rejected. Tables inside an image are returned as positioned table_cell blocks. Processing is asynchronous and this call polls until done or times out after about 75 seconds. This Worker does not store uploaded files or results. Upstream MinerU task data is temporary (≤24h). Billing: charged before the MinerU job is created (default 1 credit per tool call). No automatic refunds. Powered by MinerU (opendatalab/MinerU). License: https://github.com/opendatalab/MinerU/blob/master/LICENSE.md
Annotations the server declares
- title: Image OCR with bounding boxes
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
pdf_to_markdown_then_translateConvert a public PDF URL to Markdown via MinerU, then translate the markdown via the public text API.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "convert": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Convert a public PDF URL to Markdown via MinerU, then translate the markdown via the public text API. Pass `file_url` (or `files` with exactly 1 PDF http(s) URL; path must end in .pdf) and required `target_lang`. Optional: `source_lang` (default auto), `include_source_markdown` (default true), `wait_seconds` (default 45, max 55). URL max 2048 chars. Returns `markdown`, `translated_markdown`, `bilingual_markdown`, `task_id`, `target_lang`. Cost: 2 credits per call (MinerU 1 + translate_text 1), charged by mcp-credit-gateway before forwarding. DOCX input is not supported this wave. Compliance: Use only on documents you own or are explicitly licensed/authorized to convert and translate. Do not upload or redistribute copyrighted material without permission. Output markdown is ephemeral and must not be abused for bulk scraping. Calls are rate-limited/metered by mcp-credit-gateway and require a valid sk_live_ key. (仅处理自有或已获授权文档;禁止滥用转换/翻译他人受版权保护材料;调用限频。)
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
separate_stemsSplit a public audio URL into vocals + accompaniment stems via Spleeter 2stems (async create + result poll).Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Split a public audio URL into vocals + accompaniment stems via Spleeter 2stems (async create + result poll). Pass `audio_url` (http(s), <= 2048 chars) to start, or `job_id` to poll. Waits up to `wait_seconds` (default 30, max 55). Returns `vocals_url`, `accompaniment_url`, and optional `zip_url` when done. Cost: 3 credits per call, charged by mcp-credit-gateway before forwarding. Compliance / SSRF: only http(s) URLs; Use only on material you own or are explicitly licensed/authorized to generate or remix. Do not generate or redistribute copyrighted music, voice likenesses, or trademarked audio without permission. Output URLs have a limited TTL and must not be abused for bulk scraping or redistribution. Calls are rate-limited/metered by mcp-credit-gateway and require a valid sk_live_ key. (仅处理自有或已获授权素材;禁止滥用生成/二创他人受版权保护的音乐或声音;输出有 TTL;调用限频。)
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
text_to_sound_effectGenerate a short sound-effect WAV from a text prompt via TangoFlux (async create + result poll).Reads text others writeNo findings
Why it is in this group
- The description's first verb is "generate": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Generate a short sound-effect WAV from a text prompt via TangoFlux (async create + result poll). Pass `prompt` (<= 2000 chars) with optional `duration` (default 10s) and `steps` (default 25); the call waits up to `wait_seconds` (default 30, max 55) and returns `job_id` if still running. Poll by calling again with only `job_id`. Returns `audio_url` when done. Cost: 2 credits per call, charged by mcp-credit-gateway before forwarding. Compliance: Use only on material you own or are explicitly licensed/authorized to generate or remix. Do not generate or redistribute copyrighted music, voice likenesses, or trademarked audio without permission. Output URLs have a limited TTL and must not be abused for bulk scraping or redistribution. Calls are rate-limited/metered by mcp-credit-gateway and require a valid sk_live_ key. (仅处理自有或已获授权素材;禁止滥用生成/二创他人受版权保护的音乐或声音;输出有 TTL;调用限频。)
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
text_to_speechSynthesize speech from text via Qwen-TTS using a short `refer_audio` sample (http(s) URL).Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Synthesize speech from text via Qwen-TTS using a short `refer_audio` sample (http(s) URL). Pass `text` (<= 5000) + `refer_audio` (+ recommended `refer_audio_text`) to start, or `job_id` to poll. voiceId library voices are not exposed this wave. Cost: 2 credits per call. Compliance: Use only on material you own or are explicitly licensed/authorized to generate or remix. Do not generate or redistribute copyrighted music, voice likenesses, or trademarked audio without permission. Output URLs have a limited TTL and must not be abused for bulk scraping or redistribution. Calls are rate-limited/metered by mcp-credit-gateway and require a valid sk_live_ key. (仅处理自有或已获授权素材;禁止滥用生成/二创他人受版权保护的音乐或声音;输出有 TTL;调用限频。)
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
transcribe_audioTranscribe a public audio URL via Faster-Whisper (async create + result poll; no status endpoint).Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Transcribe a public audio URL via Faster-Whisper (async create + result poll; no status endpoint). Pass `audio_url` (http(s)) to start or `job_id` to poll. Optional `language`, `response_format` (simple|raw). Empty transcript is still success (no speech detected). Cost: 2 credits per call. Compliance / SSRF: Use only on material you own or are explicitly licensed/authorized to generate or remix. Do not generate or redistribute copyrighted music, voice likenesses, or trademarked audio without permission. Output URLs have a limited TTL and must not be abused for bulk scraping or redistribution. Calls are rate-limited/metered by mcp-credit-gateway and require a valid sk_live_ key. (仅处理自有或已获授权素材;禁止滥用生成/二创他人受版权保护的音乐或声音;输出有 TTL;调用限频。)
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
upscale_imageUpscale an image from a public URL.Reads text others write1 finding
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Warn
The description names a tool "enhance_face" that this server does not list.
What this check looks forMatched text
description: ".... There is no separate enhance_photo or enhance_face tool. Synchronous. Cost: 2 credits per call;..."
Description, as the server wrote it Upscale an image from a public URL. POST {img, upscale} to ai.imgkits.com/api/restoration. `scale` is 2 (default) or 4 and is sent as `upscale`. Optional `face_enhance` is sent on the same body. There is no separate enhance_photo or enhance_face tool. Synchronous. Cost: 2 credits per call; mcp-credit-gateway charges before forwarding.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
Only reads 2Show 2 read-only tools
count_textCount characters, words, lines, and UTF-8 bytes for `text` (≤8000 chars).No findings
Why it is in this group
- The name "count_text" says "count": a word for reading.
Description, as the server wrote it Count characters, words, lines, and UTF-8 bytes for `text` (≤8000 chars). Latin words + CJK ideographs counted separately; pure algorithm, no upstream. Returns characters, characters_no_whitespace, words, lines, bytes_utf8. Cost: 1 credit per call, charged by mcp-credit-gateway before forwarding. Privacy: Text is processed in-memory for this request only and is not stored. Do not send secrets, passwords, or personal data you would not paste into a public tool. Calls are rate-limited/metered by mcp-credit-gateway and require a valid sk_live_ key. (文本仅在本请求内存中处理、不落盘;请勿提交密钥/密码等敏感信息;调用限频。)
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
decode_blind_watermarkRead a blind watermark from an image (public http(s) URL).Reads text others writeNo findings
Why it is in this group
- The description's first verb is "read": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- untrusted-inputThe description's first sentence says "read" and "http": it reads web pages, which anyone can write.
Description, as the server wrote it Read a blind watermark from an image (public http(s) URL). Uses the AiController CreateDecodeBlindWatermarkTask + GetDecodeBlindWatermarkResult async path. Start with `source_url`; the call waits up to `wait_seconds` (default 15, max 60) and returns a `job_id` if still running. Poll by calling again with only `job_id`; polling never creates a new upstream decode task. Returns `content` when readable. Cost: 2 credits per call, charged by mcp-credit-gateway before forwarding.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write18
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out4
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 16, 2026, 14:22 UTC, unchanged since Nov 9, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 19
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 19 tools that change things (4 irreversible, 15 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://mcp.imgkits.com/mcp: tools/list returned 21 tools to an anonymous client, including translate_text (irreversible), translate_document (irreversible), translate_image (irreversible), translate_subtitle (irreversible), convert_to_markdown (write) and 14 more
Posture changes
- Nov 9, 2026
- Tools that change things, listed without credentials: 11 to 19
Findings
Matches of fixed checks, with the text they matched.
The server marks "convert_text_style" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "convert_text_style" says "convert": a word for a change.
- Warnupscale_image
The description names a tool "enhance_face" that this server does not list.
What this check looks forMatched text
description: ".... There is no separate enhance_photo or enhance_face tool. Synchronous. Cost: 2 credits per call;..."
History
Changes to its tool definitions and to its remote endpoint, newest first.
PostureRemote posture changed. Tools that change things, listed without credentials: 11 -> 19.View diff
- Tools that change things, listed without credentials: 11 to 19
Tools11 added, 2 removed.+11 added−2 removedView diff
- Added:clone_voice_ttsChanges things
- Added:colorize_photoChanges things
- Added:convert_text_styleChanges things
- Added:count_textOnly reads
- Added:generate_subtitlesChanges things
- Added:pdf_to_markdown_then_translateChanges things
- Added:separate_stemsChanges things
- Added:text_to_sound_effectChanges things
- Added:text_to_speechChanges things
- Added:transcribe_audioChanges things
- Added:upscale_imageChanges things
- Removed:add_watermark
- Removed:remove_watermark
Install
Pinned to version 1.0.1. Values in angle brackets are yours to fill in.
claude mcp add --transport http imgkits https://mcp.imgkits.com/mcp --header 'Authorization: <your-authorization>'claude mcp add --transport http imgkits https://mcp.imgkits.com/mcp --header 'Authorization: <your-authorization>'Run in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__imgkits__decode_blind_watermark",
"mcp__imgkits__count_text"
],
"ask": [
"mcp__imgkits__convert_to_markdown",
"mcp__imgkits__ocr_image",
"mcp__imgkits__extract_tables",
"mcp__imgkits__translate_text",
"mcp__imgkits__translate_document",
"mcp__imgkits__translate_image",
"mcp__imgkits__translate_subtitle",
"mcp__imgkits__detect_language",
"mcp__imgkits__embed_blind_watermark",
"mcp__imgkits__text_to_sound_effect",
"mcp__imgkits__separate_stems",
"mcp__imgkits__transcribe_audio",
"mcp__imgkits__generate_subtitles",
"mcp__imgkits__text_to_speech",
"mcp__imgkits__clone_voice_tts",
"mcp__imgkits__pdf_to_markdown_then_translate",
"mcp__imgkits__convert_text_style",
"mcp__imgkits__colorize_photo",
"mcp__imgkits__upscale_image"
],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as imgkits. Review it before you use it.
Capability tags behind 18 rules
- mcp__imgkits__convert_to_markdown untrusted-input
- mcp__imgkits__ocr_image untrusted-input
- mcp__imgkits__extract_tables untrusted-input
- mcp__imgkits__translate_text untrusted-input, outbound
- mcp__imgkits__translate_document untrusted-input, outbound
- mcp__imgkits__translate_image untrusted-input, outbound
- mcp__imgkits__translate_subtitle untrusted-input, outbound
- mcp__imgkits__embed_blind_watermark untrusted-input
- mcp__imgkits__decode_blind_watermark untrusted-input
- mcp__imgkits__text_to_sound_effect untrusted-input
- mcp__imgkits__separate_stems untrusted-input
- mcp__imgkits__transcribe_audio untrusted-input
- mcp__imgkits__generate_subtitles untrusted-input
- mcp__imgkits__text_to_speech untrusted-input
- mcp__imgkits__clone_voice_tts untrusted-input
- mcp__imgkits__pdf_to_markdown_then_translate untrusted-input
- mcp__imgkits__colorize_photo untrusted-input
- mcp__imgkits__upscale_image untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/io.github.yichong1/mcp-imgkits/verify?hash=91dda501447e10dc2bd4ca72288884b25f6d74beddfeae27e559fad67d1104e9" | jq -e '.changed == false'Pinned hash 91dda501447e, from Nov 19, 2026, 01:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.yichong1/mcp-imgkits/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.yichong1/mcp-imgkits)Source: MCP Registry · repository. Synced Nov 5, 2026, 06:17 UTC.
ImgKits MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the ImgKits MCP server safe?
- No scan can promise that. What its definitions show: 21 tools, 19 of which can change, send or delete something, and 2 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the ImgKits MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the ImgKits MCP server do things that cannot be undone?
- Yes. 4 tools can send, pay, publish or merge: translate_text, translate_document, translate_image and translate_subtitle.
What tools does the ImgKits MCP server have?
- 21 tools: 4 irreversible, 15 write, 2 read. translate_text, translate_document, translate_image, translate_subtitle, convert_to_markdown, ocr_image, extract_tables, detect_language and 13 more.
Does the ImgKits MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the ImgKits MCP server send data to outside services?
- 4 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): translate_text, translate_document, translate_image and translate_subtitle.
Has the ImgKits MCP server changed its tools?
- Yes, once since Anlyon Radar started tracking it, most recently on Nov 5, 2026. Every change is on this page and in its Atom feed.