AMZ Vault

Video: strategies, presets, reference replication (beta)

video_strategies

Read-only

Video Director creative knowledge: action=concept (assetIds or projectId, hint = the user's words, platform, level, genres[], strategyId, family: ONE committed plan with ready-to-apply scene operations, title slots, music and sound; use it when the brief is thin and build it without asking), cuts (assetId, marker, minPauseMs: a raw talking-head take becomes the cut list and ready scenes; show it as a table, edit in words, then build), preflight (projectId, promisedSeconds, intent[], expectText, expectMusic: fails and warnings with a fix each before any preview), list (built-in strategies for Amazon SBV, listing videos, A+ video, product explainer/showcase/features-benefits/demo/UGC/comparison/unboxing, social reels/shorts/LinkedIn, and story pieces; every one works with photos, clips or both; filter family=, platform=, query=), get (strategyId), brief (strategyId, or custom={label, durationSeconds, beats[...]} for a plan of your own; durationSeconds, aspectId: frame-accurate beats + presets + checklist), presets (kind=), platforms, analyze_reference (assetId: cuts, pacing, nearest strategy), replicate (assetId, optional strategyId/durationSeconds/aspectId: slot-by-slot plan to rebuild a reference with the user's footage), directions (strategyId, level safe|bold|wild, count, genres[], exclude[], seed, accent: distinct creative kits, one per video, each with engine-ready text styles, transitions, camera moves, grade, caption style, end card and a signature device; bold is the default when the user names no style). Video Director is in BETA (2026-10-18): open to every AMZ Vault account; free video accounts get 3 badged 720p exports, trials 2 clean exports a day, subscribers unlimited. Call video_guide first for the catalog, the exact input schema of each action and this account's tier notes.

Input Schema

Name	Required	Description	Default
`hint`	No
`kind`	No
`seed`	No
`count`	No
`level`	No
`query`	No
`accent`	No
`action`	Yes
`custom`	No
`family`	No
`genres`	No
`intent`	No
`assetId`	No
`exclude`	No
`aspectId`	No
`assetIds`	No
`platform`	No
`projectId`	No
`threshold`	No
`expectText`	No
`strategyId`	No
`expectMusic`	No
`durationSeconds`	No
`promisedSeconds`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 27, 2026

changedInput schema / properties / action / enum

Previous value: -[
-  "analyze_reference",
-  "brief",
-  "concept",
-  "directions",
-  "get",
-  "list",
-  "platforms",
-  "preflight",
-  "presets",
-  "replicate"
-]New value: +[
+  "analyze_reference",
+  "brief",
+  "concept",
+  "cuts",
+  "directions",
+  "get",
+  "list",
+  "platforms",
+  "preflight",
+  "presets",
+  "replicate"
+]

Changed10 schema fields changedOct 26, 2026

changedInput schema / properties / action / enum

Previous value: -[
-  "analyze_reference",
-  "brief",
-  "directions",
-  "get",
-  "list",
-  "platforms",
-  "presets",
-  "replicate"
-]New value: +[
+  "analyze_reference",
+  "brief",
+  "concept",
+  "directions",
+  "get",
+  "list",
+  "platforms",
+  "preflight",
+  "presets",
+  "replicate"
+]

addedInput schema / properties / assetIds

Added value: +{
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}

addedInput schema / properties / custom
```
Added value: +{
+  "type": "object"
+}
```
addedInput schema / properties / expectMusic
```
Added value: +{
+  "type": "boolean"
+}
```
addedInput schema / properties / expectText
```
Added value: +{
+  "type": "boolean"
+}
```

changedInput schema / properties / family / enum

Previous value: -[
-  "amazon",
-  "product",
-  "social"
-]New value: +[
+  "amazon",
+  "product",
+  "social",
+  "story"
+]

addedInput schema / properties / hint
```
Added value: +{
+  "type": "string"
+}
```

addedInput schema / properties / intent

Added value: +{
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}

addedInput schema / properties / projectId
```
Added value: +{
+  "type": "string"
+}
```
addedInput schema / properties / promisedSeconds
```
Added value: +{
+  "type": "number"
+}
```

AddedOct 15, 2026

TDQS

A3.8/5.0

Behavior4/5

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

Annotations already declare readOnlyHint=true and destructiveHint=false, and the description is consistent with pure plan/knowledge generation (outputs are 'ready-to-apply scene operations', 'frame-accurate beats', 'slot-by-slot plan'). Beyond annotations it adds genuinely useful behavior: BETA status dated 2026-10-18, tier-specific export limits ('free video accounts get 3 badged 720p exports, trials 2 clean exports a day, subscribers unlimited'), and a pointer to video_guide for tier notes.

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

Conciseness2/5

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

The description is a single dense paragraph crammed with semicolon-separated action clauses, making it hard to scan. While most content is substantive given 11 actions, the structure is poor — there is no line-per-action formatting or front-loading of the most important routing decision, and the beta/quota block is buried at the end.

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

Completeness4/5

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

For a 24-parameter, 11-action tool with no output schema and no parameter descriptions, the description supplies workflows, per-action parameter mapping, beta/quota context, and a delegation note to video_guide for exact schemas. Remaining gaps (a couple of undocumented parameters, no explicit error/return-shape discussion) are minor given the pointer to video_guide.

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 description coverage is 0%, so the description carries the full burden, and it mostly succeeds: it maps parameters to actions (concept takes assetIds/projectId/hint/platform/level/genres/strategyId/family, cuts takes assetId/marker/minPauseMs, preflight takes projectId/promisedSeconds/intent/expectText/expectMusic, directions takes level safe|bold|wild/count/genres/exclude/seed/accent, etc.). A few parameters (notably threshold, and precise semantics of kind/custom) are not explained, leaving small gaps.

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

Purpose4/5

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

The description names the tool's domain ('Video Director creative knowledge') and then enumerates all 11 actions with a specific verb+resource for each (concept, cuts, preflight, list, get, brief, presets, platforms, analyze_reference, replicate, directions). This is enough to distinguish it from siblings like video_guide or video_producer, though the umbrella phrase 'creative knowledge' is a bit vague on its own.

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

Usage Guidelines4/5

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

It gives explicit when-to-use cues for key actions — 'use it when the brief is thin and build it without asking' for concept, 'show it as a table, edit in words, then build' for cuts, and 'fails and warnings with a fix each before any preview' for preflight — plus workflow ordering ('Call video_guide first'). It lacks explicit exclusions against sibling tools other than the video_guide pointer, but the per-action guidance is unusually concrete.

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

All AMZ Vault 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?