AdaptlyPost

Upload From Device

open_upload_widget

Let the user add photos, videos or documents from their phone or computer. Use it when the user wants to post a file that has no public URL, such as a photo on their phone, a file attached in the chat that upload_media cannot read (in ChatGPT, pass attached files to upload_media as attachments instead), or a file you built in a sandbox whose network blocks the get_upload_urls uploadUrl. Never suggest putting the files in a public GitHub repo or a file-sharing link instead. In apps that show MCP Apps widgets, such as ChatGPT and Claude chat, it shows an upload box in the chat, and when the upload finishes the box sends a message listing the public media URLs. In apps that cannot show the box, such as Claude Cowork and Claude Code, it returns status upload_link with uploadUrl, a page the user opens in their browser to upload, and uploadId: give the user the link, wait until they say they finished, then call this tool again with that uploadId to get the media URLs. Pass the media URLs as mediaUrls to create_post, update_post or bulk_schedule_posts. Nothing is uploaded until the user picks files. Accepts JPEG, PNG, WebP, GIF, MP4, QuickTime, PDF, PPT, PPTX, DOC and DOCX, up to 250 MB per file, 50 MB per image and 100 MB per document. Uploaded files are stored at public URLs.

Input Schema

Name	Required	Description	Default
`uploadId`	No	uploadId from an earlier call that returned an upload link. Returns the files the user uploaded through that link instead of opening a new upload.
`workspaceId`	No	Workspace to act in: an id from list_workspaces. Omit to act in the workspace list_workspaces marks current. Use the same workspaceId for every call about the same workspace, since ids from one workspace (accounts, posts, uploads) do not exist in another

Output Schema

Name	Required	Description	Default
`result`	No	Without uploadId: status waiting_for_upload when the upload box is shown (the media URLs arrive later in a message from the box), or status upload_link with uploadUrl, uploadId and acceptsUntil. With uploadId: status (waiting, saving, ready, failed or expired), files ({ fileName, publicUrl } each), mediaUrls, failed ({ fileName, error } each), savingCount and acceptsUntil.

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedOct 31, 2026

addedInput schema / properties / uploadId

Added value: +{
+  "description": "uploadId from an earlier call that returned an upload link. Returns the files the user uploaded through that link instead of opening a new upload.",
+  "minLength": 1,
+  "type": "string"
+}

changedOutput schema / properties / result / description

Previous value: -"An object with status waiting_for_upload. The media URLs arrive later, in a message the upload box sends when the user finishes."New value: +"Without uploadId: status waiting_for_upload when the upload box is shown (the media URLs arrive later in a message from the box), or status upload_link with uploadUrl, uploadId and acceptsUntil. With uploadId: status (waiting, saving, ready, failed or expired), files ({ fileName, publicUrl } each), mediaUrls, failed ({ fileName, error } each), savingCount and acceptsUntil."

AddedSep 28, 2026

TDQS

A4.8/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare non-destructive/non-readonly; the description carries everything else: two distinct execution modes (widget vs. upload_link + uploadId round trip), 'nothing is uploaded until the user picks files', supported MIME types, per-file/type size limits, and that results are stored at public URLs. This is behavior an agent cannot get from the annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with purpose and the routing decision before mode/format details, and every sentence carries operational content. It is long, but the length is driven by two execution modes and a size-limit table rather than filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists so return values needn't be described, yet the description still explains both return paths (message with public URLs vs. status upload_link) and closes the loop by telling the agent to pass mediaUrls to create_post/update_post/bulk_schedule_posts. Nothing needed to invoke correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3; the description earns above that by explaining the uploadId two-step protocol (call again with the uploadId to retrieve files) that the schema only hints at. workspaceId is left entirely to the schema, which is acceptable given full coverage.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb+resource ('Let the user add photos, videos or documents from their phone or computer') and immediately distinguishes itself from upload_media and get_upload_urls by naming the exact conditions where those siblings fail.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly enumerates when to use it (phone photo, chat attachment upload_media can't read, sandbox file behind blocked network), names the alternative and the exception ('in ChatGPT, pass attached files to upload_media as attachments instead'), and adds a hard exclusion ('Never suggest putting the files in a public GitHub repo').

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All AdaptlyPost tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?