Skip to main content
Glama

Short-form video ads for a market (Higgsfield-style presets)

video_studio
Read-onlyIdempotent

Plan AI product videos for one market and platform, Higgsfield-style: preset formats (hero orbit, texture macro, unboxing POV, 3-step routine, local lifestyle, AI presenter) with camera moves, a shot list, keyframe and image-to-video prompts, ready request bodies for treg.to video models (draft Seedance 1 Lite, standard Hailuo 2.3, premium Veo 3.1 Fast; HeyGen avatar for presenters), hook variants for A/B tests, the platform spec (aspect, length, safe zone), voice-over language, music, AI-label and claims rules, and the estimated cost. Starts from the real packshot so the pack never changes. It reads the product's rules in that market first: sale gates to clear before spending on ads (sale_gate, e.g. a US sunscreen is an OTC drug), the rules that bind ads there (ad_rules, with official sources) and wording in the product name, description, company record or your copy that does not fit them (claims_check: what to say instead), plus local_words for captions and hashtags; signed in, it uses the company's brand, valid certifications, origin and proof points (company). read_market_first lists the market signals to read before writing hooks (this week's rising hashtags and top videos on that platform in that country, competitors' long-running ads, the best sellers' 1–3★ reviews), and each hook angle says which signal sets its first 2 seconds (from_signal). Signed in, each read_market_first step carries last_read when the company already saved that read (company_context kind=signal, with its age and whether it is stale), and each hook angle carries the company's own result there (your_views, your_hold_rate past 2 seconds; all of them in hook_tests, from company_context kind=result). Free, read-only — generation runs on the user's treg token.

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": {
      "country": "<country>"
    },
    "name": "video_studio"
  }
}

Input Schema

NameRequiredDescriptionDefault
copyNoOptional: your own ad copy, captions, voice-over script or listing text for this market; claims_check flags wording that does not fit the market's rules and says what to use instead.
itemNoProduct as an HS4 code (e.g. '3304') or a name in English or Korean (e.g. 'sunscreen', '라면'). Names cover TradeGraph's 35 items; any other product works as a 4-digit HS code (e.g. '1905' biscuits, '2005' kimchi, '3305' shampoo) with live UN Comtrade imports. Optional; sets the product group for scenes and claims rules.
tierNoModel tier: draft (cheap 480p tests), standard (Hailuo 2.3 768P), premium (Veo 3.1 Fast 1080p).
presetNoVideo format; auto = three presets that suit the product group.
traitsNoOptional: product traits, comma-separated, that change the rules within one HS code: sunscreen, whitening (also arbutin, kojic acid), vitamin_a (retinol), gel_nail (UV/LED gel, semi-cured strips), contact_lens (HS 9001: contact lenses, colour lenses), hair_dye, hair_loss, acne, antiperspirant, dandruff, fluoride (toothpaste), human_derived (human stem cell culture media, human placenta, cord blood: a ban in the EU, UK, China and ASEAN), stem_cell (stem cell culture media whose source is not named: check it), exosome, radio (Bluetooth, Wi-Fi, cellular), battery, meat, dairy, egg, aerosol, flammable, industrial or household (for HS codes that hold both, such as 8414 fans, compressors and vacuum pumps: picks machinery rules, trade fairs and buyer searches, or household-appliance ones). Also detected from item, product and description; the answer lists them in product_traits.
countryYesTarget country as ISO 3166-1 alpha-2 (e.g. 'US', 'JP') or an English/Korean name.
productNoThe product in plain English as it looks, e.g. 'vegan sunscreen SPF50, 50 ml white tube with green cap'. Used in every prompt.
platformNoWhere the video runs; auto = the market's first creator platform.
variantsNoHook variants per preset (A/B test the first 2 seconds).
descriptionNoOptional: the product as the user describes it, any language (e.g. '비건 선크림 SPF50', '할랄 인증 저가 립스틱', 'premium anti-aging ampoule already on Amazon US'). Used to detect product traits that change the rules within one HS code (see traits) and the concepts, price tier, certifications and markets already selling that change which market fits (product_profile, entry_paths).
has_packshotNoWhether a real product photo exists (default true). Without one the plan asks for it first.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / description / description
      Previous value: -"Optional: the product as the user describes it, any language (e.g. '비건 선크림 SPF50', 'Bluetooth speaker with a rechargeable battery', 'ramen with beef broth'). Used only to detect product traits that change the rules within one HS code (see traits)."New value: +"Optional: the product as the user describes it, any language (e.g. '비건 선크림 SPF50', '할랄 인증 저가 립스틱', 'premium anti-aging ampoule already on Amazon US'). Used to detect product traits that change the rules within one HS code (see traits) and the concepts, price tier, certifications and markets already selling that change which market fits (product_profile, entry_paths)."
  2. Changed3 schema fields changed
    • addedInput schema / properties / copy
      Added value: +{
      +  "description": "Optional: your own ad copy, captions, voice-over script or listing text for this market; claims_check flags wording that does not fit the market's rules and says what to use instead.",
      +  "type": "string"
      +}
    • addedInput schema / properties / description
      Added value: +{
      +  "description": "Optional: the product as the user describes it, any language (e.g. '비건 선크림 SPF50', 'Bluetooth speaker with a rechargeable battery', 'ramen with beef broth'). Used only to detect product traits that change the rules within one HS code (see traits).",
      +  "type": "string"
      +}
    • addedInput schema / properties / traits
      Added value: +{
      +  "description": "Optional: product traits, comma-separated, that change the rules within one HS code: sunscreen, whitening (also arbutin, kojic acid), vitamin_a (retinol), gel_nail (UV/LED gel, semi-cured strips), contact_lens (HS 9001: contact lenses, colour lenses), hair_dye, hair_loss, acne, antiperspirant, dandruff, fluoride (toothpaste), human_derived (human stem cell culture media, human placenta, cord blood: a ban in the EU, UK, China and ASEAN), stem_cell (stem cell culture media whose source is not named: check it), exosome, radio (Bluetooth, Wi-Fi, cellular), battery, meat, dairy, egg, aerosol, flammable, industrial or household (for HS codes that hold both, such as 8414 fans, compressors and vacuum pumps: picks machinery rules, trade fairs and buyer searches, or household-appliance ones). Also detected from item, product and description; the answer lists them in product_traits.",
      +  "type": "string"
      +}
  3. First observed

TDQS

A4.2/5.0
Behavior5/5

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

Annotations only declare it read-only, idempotent and closed-world; the description adds substantial context beyond that: free planning versus token-billed generation, richer results when signed in (company brand, certifications, origin, proof points, saved signals with staleness, past hook performance), and the fallback when no packshot exists. It also discloses that results vary by market rules and that closing sale_gate gates matters before spending on ads.

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

Conciseness3/5

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

The opening sentence front-loads purpose well, but the body becomes a long semicolon-chained run that mixes outputs, market-rule sources, signed-in enrichment and signal provenance into two dense paragraphs. Much of it earns its place, yet the packing makes key behaviors (cost, packshot prerequisite, auth effects) easy to miss.

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 an 11-parameter planning tool with no output schema, the description is unusually complete: it names the returned artifacts, the rule sources consulted (sale_gate, ad_rules, claims_check, local_words), how hooks tie to signals, and the cost estimate. Only the shape of the returned payload remains loosely specified.

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

Parameters3/5

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

Schema coverage is 100%, so the schema already documents every parameter including enum meanings for tier, preset, platform and the long traits list. The description mostly restates this at the feature level (claims_check, product_traits, from_signal, hook_tests) rather than adding syntax or format meaning per parameter, so the baseline 3 applies.

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 and resource (plan AI product videos for one market and platform) and enumerates the concrete artifacts returned: preset formats, camera moves, shot list, keyframe/image-to-video prompts, ready request bodies for named video models, hook variants, platform spec and cost. This is clearly distinguishable from siblings like image_studio (stills) and channel_plan (channel selection).

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?

Gives a clear operating sequence — read the market's rules and signals first, then write hooks — and notes the cost model ('Free, read-only — generation runs on the user's treg token'), plus the packshot prerequisite and what happens when it is missing. It never explicitly names a sibling alternative or an exclusion condition, so it falls short of the top band.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.