Agents
The Sume video agent from code: Agent Completions, scheduled runs, MCP tools and timeouts, programmatic tool calling, and guardrails for paid APIs.
Start with What is a video agent? How Sume defines and runs one
Can Claude make videos? Only by calling a video tool
Claude's output is text, not video. It can still get you a clip by calling a video tool, such as Sume's hosted MCP server added as a connector.
Can Claude generate images? No, but it can call a tool
No: Anthropic says Claude understands images but cannot create or edit them. Connect an image tool over MCP, such as Sume's generate_image.
Can Claude edit videos? Only by calling editing tools
Claude takes text and images, not video, and replies in text. It can still cut, crop, or assemble a clip by calling editing tools over MCP.
Video generation MCP server: how Sume's generate_video works
Sume's hosted MCP server has a paid generate_video tool: a prompt or image in, a job id back, then jobs_wait and jobs_result for the clip.
Agent skills vs MCP: what each does and when to use both
An agent skill is know-how the agent reads when a task matches; an MCP tool is an action it calls on a server. How they differ, and when to use both.
Human in the loop AI agents: where a person approves
Human in the loop means an AI agent stops for a person before steps that matter. How Sume's agent does it in chat, and what replaces it in API runs.
A Claude skill to generate video: what works today
Anthropic's pre-built skills cover documents, not video. A Sume CLI skill lets Claude Code make avatar videos; other clips need MCP or the API.
MCP server for image editing: restyle, cut out, upscale
Sume's hosted MCP server gives agents three image edits: prompt edits from reference images, background removal to a PNG with alpha, and upscales.
MCP prompt injection: limit what a hijacked agent can do
MCP prompt injection gets an agent to call tools nobody asked for. Bound the session instead: fewer tools, read-only by default, confirmed writes.
Difference between an MCP server and an AI agent
An MCP server exposes tools; an agent is the program where a model decides which tools to call. How the work splits, with Sume's server and agent.
YouTube chapter generator: timestamps from a transcript
A YouTube chapter generator turns a transcript into timestamps and titles from 00:00: sentence timings from speech-to-text, topic breaks from an LLM.
Transcribe audio to text with Claude using an MCP tool
To transcribe audio to text with Claude, give it a speech-to-text tool over MCP and a public audio URL, then read the transcript and word timings.
Claude text to speech: make audio files with an MCP tool
For text to speech in Claude, connect a text-to-speech tool over MCP: Claude sends the script and a voice, then hands back a link to the audio file.
Make AI UGC videos with Claude: avatar, voice, lip sync
Claude's models output text. Connect Sume's hosted MCP server with Write on, and Claude can script, voice, and lip-sync a UGC-style avatar clip.
AI image description generator: alt text and captions by API
Send an image to an agent API that can see it and ask for alt text, a title, or a product description as JSON fields. How to run it with Sume.
How to make a video ad with AI: brief, draft, approve, reuse
To make a video ad with AI, brief a video agent on the deliverable, review a draft before the paid render, then save the recipe to reuse it.
Generate images for a website with Claude Code and Sume MCP
Connect Claude Code to Sume's hosted MCP with Write access, preview the cost with dry_run, then generate site images and save them in your repo.
Summarize a video with an API: transcript, stills, then JSON
Summarize a Sume-hosted video: pull stills and a transcript with POST /v1/video-inspect, then send both to Agent Completions with an output_schema.
What is Sume? A video agent platform, its API, and billing
Sume is a video agent platform: brief an agent in chat, save the recipe as a Format, and call it from your backend over one API. Surfaces and billing.
Sume Agent Completions vs Format vs Scheduled runs: request body diff
Sume Format runs, Scheduled runs, and Agent Completions share field names, not rules: spend-cap defaults, null, on_active_run, attachments, and scopes differ.
Prevent overlapping AI agent runs on Sume with on_active_run
Sume's on_active_run field decides what a second run request does while one is in flight: run anyway, record a skipped run, or get a 409 and nothing runs.
MCP vs CLI vs API for AI agents: which Sume interface to use
On Sume, HTTP and the Format API are the primary path for AI agents. Hosted MCP fits remote MCP clients; the CLI fits shell agents and is Avatar-heavy.
MCP tool call timeouts on long-running video jobs: use jobs_wait
On Sume's hosted MCP server, one jobs_wait call holds at most 55 seconds. Wait out long video jobs in slices, batch up to 20 ids, and never resubmit.
Sume MCP tools list: hosted tools grouped by read, write, and paid
The Sume MCP tools list by family: which hosted tools only read, which write, which start paid jobs, and which tool names are not on hosted MCP at all.
Programmatic tool calling on MCP: how Sume's script_run works
script_run is programmatic tool calling on Sume's hosted MCP: a short JavaScript program calls tools in loops or in parallel under call and paid budgets.
Video editing MCP server: let an AI agent trim, filter, and assemble
Sume's hosted MCP server gives AI agents video editing tools: inspect, frames, trim, filter, audio detach, compose, and a Timeline render, each run as a job.
List scheduled agent runs over the Sume API: audit your schedules
List Sume schedules with GET /v1/actions (limit, status, trigger_type), page each run history with a cursor, and read trigger.source to tell cron from API runs.
Agent API with image input and JSON output: Sume Agent Completions
Send images to Sume Agent Completions as input_image parts or attachments, bind an output_schema, and read typed JSON from output when the run completes.
What is a video agent? How Sume defines and runs one
In Sume's docs, a video agent is a sandbox Agent that composes generation tools into a post-ready video. Brief it in chat, or call it over HTTP.
Safe automation for AI agents that call paid APIs
Keep agents read-only by default, keep secrets out of logs, and on hosted MCP send an idempotency_key, preview with dry_run, and cap with max_spend_usd.
Scheduled AI video agent runs: cron, API triggers, and receipts
A Sume schedule is a saved Agents automation that runs on a cron cadence and returns a run receipt. Author it in the dashboard; start and monitor runs by API.
Run the Sume video agent from your backend with Agent Completions
POST /v1/agent/completions runs the same agent as the Sume Agents chat, with tools and media generation, and returns an async run receipt you poll or webhook.