Skip to main content
Glama
Upload-Post

Upload-Post

Official

Manage YouTube live

manage_youtube_live
Destructive

Create, update, bind, transition, and delete YouTube live broadcasts and streams, plus send chat messages, manage polls, and insert ad cuepoints. Preview and confirm actions before changes.

Instructions

Live broadcasts of the connected channel. Upload-Post does not send video: it returns the ingestion URL and stream key for the user's encoder (the key is a secret; share it only with the user). create_broadcast (title, scheduled_start_time, privacy_status, description, made_for_kids, enable_auto_start, enable_auto_stop, latency_preference…; create_stream true also creates and binds a stream), update_broadcast, delete_broadcast (needs confirm); bind (stream_id, or unbind true); transition (status testing|live|complete; live and complete need confirm); insert_cuepoint (duration_secs: an ad break); create_stream (title, ingestion_type, resolution, frame_rate), update_stream, delete_stream (needs confirm; encoders using its key stop); live chat of the channel's own broadcasts: send_chat_message (broadcast_id or live_chat_id, message up to 200 characters or poll {question, options}; daily limit per channel), close_poll (message_id), delete_chat_message (message_id, needs confirm). Every call spends the account's daily YouTube allowance (free plan 200 units, paid plans 2,000; resets at midnight Pacific time; youtube_read quota shows what is left): reads cost about 1 unit, most writes 50, captions 400-450, so batch reads and avoid needless writes. Needs the user's approval: call it first without confirmed to get a preview (nothing is sent), show the user exactly what will go out, and repeat the call with confirmed: true only after they approve.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "action": "<action>",
      "user": "<user>"
    },
    "name": "manage_youtube_live"
  }
}

Input Schema

NameRequiredDescriptionDefault
userYesUpload-Post profile name with YouTube connected.
actionYes
paramsNoOther API fields for this action, with the REST API's own names (snake_case) and shapes. See the description for each action's fields. Leave out what you do not want to change.
confirmNoThe API's own confirmation for deletions and other changes that cannot be undone. Set it to true only after the user has approved that exact change; without it those actions are refused and nothing changes.
confirmedNoSet to true only after the user has seen the preview this tool returns without it and approved it. Without it nothing is sent.
stream_idNoupdate_stream / delete_stream, or the stream to bind.
message_idNoclose_poll / delete_chat_message: the chat message.
broadcast_idNoThe broadcast (update, delete, bind, transition, insert_cuepoint; or the chat to write to).

Output Schema

NameRequiredDescriptionDefault
resultNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv0.25.0

TDQS

A4.7/5.0
Behavior4/5

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

Annotations already declare destructiveHint=true and openWorldHint=true, and the description adds substantial context on top: which actions are irreversible and require `confirm`, that deleting a stream stops encoders using its key, that the stream key is a secret to be shared only with the user, and the two-step preview/approve flow that sends nothing on the first call. The quota model (free 200 units vs paid 2,000, reset at midnight Pacific, reads ~1 unit, writes ~50, captions 400-450, per-channel daily chat limit) is real behavioral detail the annotations cannot express. What it does not cover is the error/refusal shape or rate-limit backoff behavior, keeping it just below a 5.

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?

For twelve actions the length is justified and almost every clause carries information; the most surprising constraint (no video upload, key is a secret) and the approval protocol are placed first and last respectively. The weakness is structural: the middle is a long semicolon-delimited run of action names, so an agent must parse prose rather than scan a list. No filler sentences, but the density makes re-reading likely.

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 need not be explained, and everything else an agent needs is present: the user profile prerequisite, all twelve actions with their fields, which actions are destructive and require `confirm`, the mandatory preview/approve protocol, and the quota economics that affect batching decisions. For a destructive, open-world, multi-action tool this is complete enough to invoke correctly without opening another document.

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

Parameters5/5

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

Schema coverage is 88%, so the bar is a baseline 3, but the description exceeds it because the `params` property is a free-form additionalProperties object that the schema cannot document at all — the description is the only place an agent learns that create_broadcast takes title, scheduled_start_time, privacy_status, made_for_kids, enable_auto_start/stop and latency_preference, that insert_cuepoint takes duration_secs, or that chat messages are capped at 200 characters with an optional poll shape. It also disambiguates the two easily-confused flags: `confirm` is the API's own confirmation for irreversible changes, `confirmed` is the preview-approval gate.

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?

The opening sentence names the resource domain (live broadcasts, streams and live chat of the connected channel) and immediately adds the non-obvious scope fact that Upload-Post does not transmit video but returns an ingestion URL and stream key. The action enum is then enumerated with the object each one operates on (broadcast, stream, cuepoint, chat message, poll), so an agent can distinguish create_broadcast from create_stream or send_chat_message at a glance. It also points at the sibling youtube_read for quota inspection.

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?

Usage conditions are stated per action: create_stream true on create_broadcast binds a stream, bind takes either stream_id or unbind, transition accepts testing|live|complete, send_chat_message takes broadcast_id or live_chat_id. It gives explicit when-not/avoid guidance for quota (batch reads, avoid needless writes) and an explicit approval protocol (call without `confirmed` first for a preview, repeat with confirmed:true only after approval).

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

Deploy Server

Other Tools