Enterprise AI sales and support agents for WhatsApp, Instagram, Telegram and web chat. Turn every inbound message into qualified revenue — 24/7.
This listing identifies a third-party MCP server. MCPBundles is not affiliated with, sponsored by, or endorsed by the service or its operator unless stated on this page.
This server may ask you to sign in when your AI client connects.
Last probed Nov 3, 2026 · mcp.mybots.pro
55tools discovered
For AI agents — MCPBundles Directory API
For MCP client apps (human connect):
MCP connect URL: https://mcp.mybots.pro/mcp
Skill page: https://www.mcpbundles.com/skills/mybots-mcp-335e44f8d2
Explore capabilities (55)
read 10
write 15
Showing 25 of 55
Showing 25 of 55 tools from the public preview. Connect the server to inspect the full live tool list.
Adobe8
Blender3
Generate2
Get2
List2
Other capabilities8
Record the user's viewport animation as a reference clip
write
blender_playblast
Renders the viewport (or scene camera) animation to a clean 1280x720 clip on the user's machine, uploads it to their Mira library and returns its URL, ready for referenceVideoUrls in generate_video (seedance-2-5, minimax, wan3, wan3-prime, kling, kling-motion, omni) - pass ratio '16:9' with it, the clip is landscape and the models default to 9:16. Keep the range short: models take 3-30 s of reference; the tool refuses more than 30 s. By default figures keep travelling and turning but their gestu
Description may be truncated on the public preview.
Record the user's viewport animation as a reference clip — Renders the viewport (or scene camera) animation to a clean 1280x720 clip on the user's machine, uploads it to their Mira library and returns its URL, ready for referenceVideoUrls in generate_video (seedance-2-5, minimax, wan3, wan3-prime, kling, kling-motion, omni) - pass ratio '16:9' with it, the clip is landscape and the models default to 9:16. Keep the range short: models take 3-30 s of reference; the tool refuses more than 30 s. By default figures keep travelling and turning but their gestu
Generate an image — Renders an image from a text prompt, optionally guided by reference photos. SPENDS THE USER'S CREDITS: 3 per image, 1 on muse-image, 4 on seedream-5-pro, 8 on the premium models (nano-banana-pro, gpt-image-2.5-sunburst). Say what it will cost before running a batch.
Write the prompt in ENGLISH and make it complete and vivid. Everything you state is kept word for word - the subject, the setting, the moment, any lettering - while a pipeline of specialists fills only the craft slots you left empty
See a frame of the active composition — Renders a frame of the active composition/sequence and returns it inline as a PNG: the one under the playhead, or the one at `time` seconds (After Effects; the playhead does not move). Check a shot, a transition at cut-0.2 / cut / cut+0.2, or use it as a reference (upload_reference_image) for generate_*.
Generate a video clip — Renders a short video clip from a text prompt, or animates a reference image. SPENDS THE USER'S CREDITS, and video is far more expensive than an image - the price depends on model, duration and resolution. Always tell the user which model you picked and that it will spend credits BEFORE calling this.
Write the whole clip prompt in English as plain cinematic prose, not hand-assembled model syntax: a pipeline of specialists (cast and blocking, scene and props, camera, light, look and sound) emits
Place a Mira Element into the user's composition — Downloads an element from list_elements on the user's machine and places it: overlays, backgrounds and transitions become a layer over the active composition at the playhead (or at `time`) with their blend mode, loop and opacity; a luma wipe (blend 'luma') becomes the track matte that reveals the incoming layer starting at `time` (or options.incoming); a sound effect becomes an audio layer whose transient lands on `time`; a template project is imported, its texts and controls set from `texts` /
Get your MCP into directories
A working endpoint is step one. Directory coverage is the coordinated launch across ChatGPT, Claude, Cursor, the MCP Registry, and community indexes.
Enterprise AI sales and support agents for WhatsApp, Instagram, Telegram and web chat. Turn every inbound message into qualified revenue — 24/7.
How do I connect myBots to my AI agent?
Use the MCP endpoint listed on this page in your MCP client configuration. One-click install pills support Claude, Cursor, VS Code, and other hosts. Copy the remote MCP URL if your client needs a manual entry.
How many tools does myBots provide?
Operate this server?
Request removal if you do not want myBots listed here, or claim the listing if you want to manage it on MCPBundles.
Operate myBots?
This server appears in the MCPBundles directory. Verify you operate it to take over the listing — name, description, logo, contact email, and skill content. We email a 6-digit code to a maintainer address your server publishes in /.well-known/security.txt or /.well-known/mcpbundles.json. Free, takes about a minute.
MCPBundles publishes this directory for MCP discovery. Except where we host or operate an offering, third-party services run under their own terms. Product and company names on this page are used in a descriptive, identifying way (including under nominative fair use where applicable); they remain the property of their owners. Nothing here grants you rights in those marks, and nothing here is an offer to sell a third party's services. Terms · Request removal
More in developer tools
Other MCP servers in this category from the directory index
Replace the person in a clip with another character — Swaps the person in a clip for the character in an image while keeping the performance, camera and background (Wan animate/replace). SPENDS CREDITS (per second of the source; see estimate_cost kind 'recast'). The clip is one of the user's Mira clips (sourceGenerationId) or an uploaded clip (libraryItemId) - pass exactly one - up to 30 s. characterImageUrl is an https URL of a Mira image or an upload. resolution: '480p', '580p' or '720p' (default). Returns a new generation_id; call wait_for_gener
Run a named panel command — Runs one of the panel's typed After Effects commands with a JSON input instead of raw ExtendScript: list_compositions {}, get_composition_info {comp|name}, list_layers {comp} (index, type, in/out, 3D, parent, blend, track matte, effects, markers), create_composition {name,width,height,fps,duration}, add_text_layer {comp,text,name}, add_solid_layer {comp,color,name}, set_keyframe {comp,layer,property,time,value}, apply_effect {comp,layer,effect}, set_track_matte {comp,layer,matteLayer,mode: luma|
Reframe a clip to another aspect ratio — Re-frames a video to a new aspect ratio (9:16, 16:9, 4:3, 3:4, 1:1, 21:9) and generatively fills the newly exposed canvas, keeping the subject centred — a vertical cut from a landscape clip without letterboxing. SPENDS CREDITS (per second of the source; see estimate_cost kind 'reframe'). The source is one of the user's Mira clips (generation_id) or a clip uploaded through create_upload (library_item_id), up to 15 s. Returns a new generation_id; call wait_for_generation.
List Mira skills (playbooks) — Lists the Mira skills bundled with this server: playbook skills whose id you can pass as the `skill` parameter of generate_image / generate_video (ugc, product, ads, cinematic, ...), and knowledge skills about prompting on this platform (mira-video-prompting, mira-image-prompting, mira-craft, mira-generate). Free, read-only. Call it once per session before you pick a skill id; to read a skill's text call get_skill.
Cancel a background render — Stops a background (aerender) render from adobe_render and deletes its partial file. A render inside After Effects cannot be cancelled from here.
Check a generation's status — Returns the current status of one generation and, once it is done, the URLs of the finished media (for transcripts and analyses the result is in files[] - role 'transcript'/'srt'/'analysis' - and urls is empty). Returns immediately - use wait_for_generation instead when you just want the result.
Talk to Mira, the platform's creative agent — Sends a message to Mira — the user's creative AI on Mira AI — in one of their dialogs and returns her reply in full. Mira knows the user's avatars, products and past generations and can start image, video and 3D generations herself (they come back as cards with generation ids). Each reply costs 2 credits. Use it when the user wants Mira's take or a brief, or to hand her a whole task; for plain generations call generate_* directly. Keep dialogId to continue the same conversation.
List the objects in the user's Blender scene — Returns the scene as data: every object with type, location, rotation, dimensions, parent, collection and visibility, plus cameras, lights, the active camera, frame range, fps and units. Use it to plan changes before writing code with blender_execute_python.
Analyse a transcript: silences, best clips, SEO, chapters, multicam — Turns a transcript from transcribe_video into an editing decision: task 'clean_cut' finds the silences to remove (cuts[{s,e}], tuned by minSilenceMs/paddingMs), 'multicam' splits the timeline by speaker (segments[{s,e,speaker}]), 'clipper' picks the best short clips for a platform (clips[{s,e,title,hook}], tuned by maxClips/clipSeconds/platform), 'seo' writes a title, description, tags, hashtags and chapters (seo{}, chapters[]), 'chapters' writes chapter markers (chapters[{s,title}]). clean_cut
Find the beat of a music layer and mark it — Analyses the music layer of an After Effects composition on the user's machine (tempo, a steady beat grid, 4/4 downbeats, drops where the bar gets louder or the kick doubles, low/mid/high sections) and writes composition markers 'mira:bar 5', 'mira:beat 12', 'mira:drop' (Mira's own markers are replaced, the user's stay). Returns bpm, grid_offset and beat_seconds (beat k is at grid_offset + k * beat_seconds in composition time), downbeats, drops and sections: cut, place transitions (whoosh 0.3-0.
List available generation models — Lists the image, video and 3D models this account can use right now, plus the accepted aspect ratios, clip durations and resolutions, and for 3D models their inputs, formats, quality tiers and credit prices. Call it when you are unsure whether a model name is valid, or when the user asks what is available. Model names are validated server-side: an unknown name silently falls back to the registry default rather than failing, so checking here is how you avoid rendering with the wrong model. An emp
Get a one-time URL to upload a clip, an audio track or a 3D mesh — Returns a one-time URL for uploading a large file that does not fit upload_reference_image: a video clip to use in referenceVideoUrls or as the libraryItemId source of the clip operations (reframe_video, modify_video, lipsync_video, transcribe_video...), an audio track (kind 'audio') for lipsync_video / change_voice, or a GLB/FBX/OBJ mesh to rig with rig_3d. Spends no credits. PUT the raw file bytes to upload_url with the returned content_type (no auth header, no multipart) within 30 minutes; th
Read a Mira skill — Returns the text of one bundled Mira skill. Without `sections` you get its SKILL.md router and the list of section keys; then call again with `sections` (comma-separated keys exactly as listed) to read the sections the router pointed you to. Free, read-only.
Import a finished generation into the user's project — Downloads a succeeded generation (image or video, from generate_image / generate_video / list_generations) into the user's Adobe app: into the Mira bin/folder, or straight onto the active composition/sequence at the playhead with destination 'active'. A remove-background result arrives with its matte: with destination 'active' it lands above the clip as its luma track matte. Returns the imported item name.
Transcribe the speech of a clip — Transcribes the speech of a clip with word timestamps and, when diarize is true (default), speaker labels. SPENDS CREDITS (per minute of the source, minimum 1; see estimate_cost kind 'transcribe'). The source is one of the user's Mira clips or audio generations (sourceGenerationId) or an uploaded clip/track (libraryItemId) - pass exactly one - up to 30 minutes. language is an optional ISO-639-1 hint. The result is a generation of kind 'transcript' with an empty urls and two files in files[]: rol
Upload a reference image — Uploads an image the user has on disk and returns a URL you can pass straight into referenceImageUrls. Spends no credits. Use it when the user points at a local file - 'use this photo', 'take the picture from my desktop' - and you can read that file yourself; send the raw bytes base64-encoded. If the picture is already on the public web, skip this and pass its URL directly. Note: an image the user merely attached to the chat is not readable this way - ask them for a file path or a link instead.
Is the user's After Effects / Premiere Pro connected? — Tells whether the user's Adobe app with the Mira panel is connected to this server and, if so, which app (aeft = After Effects, ppro = Premiere Pro, phxs = Photoshop), its version, the open project, the active composition/sequence and the selection. `busy` says what the panel is running right now (a long script, a render): other adobe_* calls wait for it in order. Call this first before any other adobe_* tool. Spends no credits.
See the user's viewport — Captures the user's 3D viewport (or the scene camera) as an image and returns it inline, so you can check the result of your changes. Overlays and gizmos are hidden in the capture.
List compositions / sequences and layers of the user's project — Returns the project as data: compositions (After Effects) or sequences (Premiere Pro) with size, fps and duration, the active one with its layers/clips (name, source file, in/out, enabled), and the playhead. Use it to plan edits before adobe_execute_script.
Wait for a generation to finish — Blocks until the generation finishes, then returns the media URLs. It never blocks for more than about a minute: if the work is still running when that budget is spent it returns status 'running' with poll_after_ms, and you should simply call it again with the same generation_id. Video routinely needs several such calls. Show the returned URLs to the user when the status is 'succeeded'. Audio comes back as an mp3 URL in urls[0]; transcripts and analyses have an empty urls and their JSON/SRT in f
Showing the first 25 tools — full list and schemas: Directory API.
MCPBundles probed 55 tools on the live server. The tool list on this page reflects what was discovered at the last refresh — connect your client to see the full set available to your session.
What authentication does myBots require?
myBots may require signing in to the provider before tools can run. Connect through MCPBundles or your MCP client and complete any provider login when prompted.
Don't want this listed?
Request removal from the MCPBundles directory. We stop publishing this listing on mcpbundles.com.
MCPBundles is an independent platform built on the open Model Context Protocol standard. MCPBundles is not affiliated with, sponsored by, or endorsed by Anthropic PBC.