async-analyze

async-analyze is a command for Claude Code from twelvelabs-io/twelve-labs-claude-code-plugin. It costs 27 tokens per session (3,186 once invoked), scanned A, original, MIT.

A command that analyzes a video in the background from a URL, local file, uploaded asset, or base64 data. It returns a task ID so you can retrieve the result later.

In plain words
What is it for?
Use it to analyze videos up to two hours long, request time-based descriptions, or include reference images in the analysis.
Why use it?
It lets you work with videos that are not already indexed and avoids waiting for long analyses to finish immediately.

Command for Claude Code

Written for Claude Code: disable-model-invocation in frontmatter.

Not installable: its command points at a path on the author’s own machine, so it runs nowhere else. The line is /Users/me/clips/keynote.mp4.

Part of the twelvelabs plugin — 10 skills, 11 commands, 2 hooks, 1 MCP server shipped together

Good fit Use it to analyze videos up to two hours long, request time-based descriptions, or include reference images in the analysis.

Compare 6 commands from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add twelvelabs-io/twelve-labs-claude-code-plugin
Claude Code
/plugin install twelvelabs

Made for: Claude Code.

Or install twelvelabs, the plugin that ships this one along with the rest of its 10 skills, 11 commands, 2 hooks, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for async-analyze

README.md
[![agentmods](https://agentmods.dev/badge/commands/twelvelabs-io/twelve-labs-claude-code-plugin/async-analyze.svg)](https://agentmods.dev/commands/twelvelabs-io/twelve-labs-claude-code-plugin/async-analyze)
Your own site
<a href="https://agentmods.dev/commands/twelvelabs-io/twelve-labs-claude-code-plugin/async-analyze"><img src="https://agentmods.dev/badge/commands/twelvelabs-io/twelve-labs-claude-code-plugin/async-analyze.svg" alt="Measured on agentmods" height="20"></a>
Per session 27 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,186 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00027 $0.03186
Opus 5 $0.00014 $0.01593
Sonnet 5 $0.00005 $0.00637
Haiku 4.5 $0.00003 $0.00319

Measured 8d ago against content hash a5ee5a0b947d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

async-analyze scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

commands/async-analyze.md · 359 lines

How it starts

The opening of the file, as written. The whole thing — 359 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/twelvelabs:async-analyze - Async Video Analysis

Asynchronously analyse a video using Pegasus 1.5 (default). Accepts a direct URL, a local file path (auto-uploaded as an asset), a previously uploaded asset ID, or base64 data — no prior indexing required. Handles videos up to 2 hours. Returns a task ID immediately; poll with status to retrieve the result.

Use this when:

  • You have a video URL, local file, or asset but no indexed videoId
  • The video is over 1 hour long (sync analyze is capped at 1 hour)
  • You want structured time-based metadata segmentation
  • You want multimodal prompting with reference images

For already-indexed videos under 1 hour, /twelvelabs:sync-analyze is faster.

Usage

/twelvelabs:async-analyze <url-or-path-or-asset-id>
/twelvelabs:async-analyze <url-or-path-or-asset-id> "<prompt>"
/twelvelabs:async-analyze status [task-id]
/twelvelabs:async-analyze list
/twelvelabs:async-analyze delete <task-id>

User provided: $ARGUMENTS

Arguments

  • First positional — one of:
    • Direct http(s) URL to a raw media file (YouTube/Drive/Dropbox share links not accepted)
    • Absolute or relative local path to a video file (this command will upload it as an asset for you)
    • A previously uploaded asset ID (24-char hex string starting with a digit)
  • prompt - Optional: natural-language question or instruction (max 2000 tokens). Drives general analysis. Cannot be combined with analysisMode: "time_based_metadata".
  • status - Check task status. Optionally provide a task ID.
  • list - List recent async analyse tasks (supports filters — see List Flow).
  • delete - Delete a task by ID.

Optional advanced parameters (when the user asks for them)

Param Purpose
modelName "pegasus1.5" (default) or "pegasus1.2" (legacy, deprecated). Set 1.2 only if the user explicitly requests it.
promptV2 Structured prompt with <@name> references to attached images (Pegasus 1.5 only). Mutually exclusive with prompt. Shape: { inputText, mediaSources: [{ name, mediaType: "image", url|assetId|base64String }] }. Up to 4 mediaSources.
temperature 0-1, default 0.2
maxTokens Pegasus 1.2: 1-4096. Pegasus 1.5: 2048-32768 (default 32768).
jsonSchema JSON Schema (Draft 2020-12) — constrain general-mode output to structured JSON. Cannot combine with segmentDefinitions.
startTime / endTime Clip the analysis window (Pegasus 1.5 only). end - start ≥ 4. Mutually exclusive with segmentDefinition.timeRanges.
analysisMode "time_based_metadata" — switches output to per-segment structured extraction. Requires segmentDefinitions. Pegasus 1.5 only.
segmentDefinitions 1-10 segment definitions; each has id, description, optional fields (typed schema), optional mediaSources (up to 4 reference images), optional timeRanges.
minSegmentDuration / maxSegmentDuration Constraints on auto-extracted segment lengths. time_based_metadata only. Mutually exclusive with per-definition timeRanges.
customId Caller-supplied tracking ID (1-64 chars, alphanumeric + -_). Round-trips through get-analyse-task and list-analyse-tasks.

Read the full file on GitHub · 359 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 359 lines · 27 tokens per session scan A a5ee5a0b947d

Subscribe to this mod's changes

async-analyze is a command published in the GitHub repository twelvelabs-io/twelve-labs-claude-code-plugin (23 stars, last pushed 3mo ago), licensed MIT. It adds 27 tokens to every session and 3,186 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.