Skip to main content
Glama
Hermoso

Hermoso

Official

Make an explainer video

make_explainer

Turn a topic or script into a narrated explainer video with selectable visual style, pacing, duration, and optional captions, returning a draft before any HD render.

Instructions

AI HOST EPISODE: format 'host_episode' = one presenter talking to camera, the camera changing every piece (frontal, three-quarter, close), the same host and set throughout, native voice, 16:9. It returns a 480p DRAFT; HD (720p) is a SEPARATE call with fromDraft (quote it with dryRun, run it only when the user asks). Otherwise: turn a TOPIC into a finished narrated explainer video, in one of TWO LANES (lane). BURNED ON-SCREEN TEXT IS OFF BY DEFAULT; the film ships clean unless the user asks; captions:true adds held key points and subtitles:true adds subtitles timed to the narration. Takes SEVERAL minutes — one image render per frame; NOT for a short product ad: use render_ad, generate_video or make_template_ad. Full rules: get_instructions({topic:'make_explainer'}).

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "make_explainer"
  }
}

Input Schema

NameRequiredDescriptionDefault
laneNo'blocks' (default) = 10-second video blocks, real motion, one narrated line per block; 'stills' = the picture film (a still about every 1.5s, narrated, no video model; cheaper). Quote either with dryRun.
musicNomusic bed under the narration, measured to sit about 14 dB under the voice and sidechain-ducked beneath it. Omit and the KIDS and FAIRYTALE channels get their recommended bed COMPOSED for this film — those two are the only channels a bed is due on unasked, and it costs a small flat fee; every other channel ships dry. 'off' forces silence. 'library' takes a free curated track only, and ships dry when none is on file. NAME A MOOD (upbeat / calm / warm / epic / tense / playful / elegant / hype / chill / dramatic) or DESCRIBE it in words to compose one on ANY channel, at the same fee. hermoso_capabilities reports the exact figure as explainerMusicCredits; quote it before you turn a bed on or pick a mood.
styleNovisual style: 'cinematic' (default, photoreal); styled shortcuts editorial_collage, flat_vector, stickman, whiteboard, ink_marker, silhouette, storybook, paper_diorama, isometric, claymation, pixel_art, watercolor, fluffy_toy, low_poly, stylized_3d, studio_3d (the Kids default), mannequin; or ANY look described in words ('80s anime cel animation'), locked across every frame. Ask rather than pick silently; a styled look costs more.
topicNowhat the explainer should teach or explain — a topic or a short brief (host_episode: or pass `script`)
voiceNonarration voice name — omit for the default warm read
dryRunNotrue = return the estimated credits this explainer reserves (its own pricing, stopped at the hold) and render nothing. Quote it before running one; try frameDensity lean or minimal when the balance is short.
formatNoblocks lane: 'explainer' (default; Gemini Omni 720p, 16:9 or 9:16, runs exactly the length asked) or 'faceless_channel' (a YouTube/TikTok faceless channel video; MiniMax H3 at 2K, 16:9 by default, five hard cuts per 10s block, runs exactly the length asked: a last part-block is rendered whole and cut). Omit and a history / kids / fairytale channel is a faceless channel video. 'host_episode' = the AI host episode (see the top).
scriptNohost_episode: the exact words, said verbatim and split at natural breaks into 4-30s pieces
camerasNohost_episode: the rotation, ids frontal / three_quarter / close or framings in words; one entry = one fixed camera
channelNothe CHANNEL TYPE — it sets the pacing, the narration register and the default look, and is orthogonal to `style` (a named style always wins): explainer (casual second-person, fast cuts), history (witty chronological retelling / documentary), kids (fastest, question-first, warm teacher), fairytale (slow, atmospheric myth or folklore). Default 'explainer'.
creatorNohost_episode: the host, a saved creator or a preset by name or id (list_creators)
endCardNoappend the branded end card. DEFAULT FALSE — set true ONLY when the user asks for one
settingNohost_episode: the set in words (default: written to fit the topic)
upscaleNooptional FINAL upscale — 2 doubles each side, 4 quadruples. Captions and the end card are burned BEFORE it so they upscale with the frame. It is priced BY LENGTH and it is the expensive part — several times the cost of rendering the film itself. hermoso_capabilities reports the exact figures per length as explainerUpscaleCredits. Never turn it on unasked: quote the number and let the user choose.
captionsNoturn ON-SCREEN TEXT on. DEFAULT FALSE, and leave it false unless the user asks — the narration already says the point and the pictures carry it, so the clean film is the better default. `captions:true` on its own burns SUBTITLES (see below), because that is what a caption is for: showing what is being said when the phone is on mute. Slim white text (CAPS in the stills lane), thin black outline, bottom safe band, no plate, no box.
fromFilmNoA PRESENTER ON A FINISHED FILM (only when the user asks for one): the job id of a finished video-blocks explainer or faceless-channel film. With `presenter`, that film comes back with a presenter speaking each block's line on screen; every block keeps its own picture and the film without the presenter is kept (cleanVideo). Never add a presenter unasked.
get_costNotrue: return the price only; nothing runs or bills
brandNameNobrand name for the end card — omit to leave it unbranded
fromDraftNohost_episode: a finished draft's job id, to render it in HD (720p) with the same script, cameras, set and host
hostImageNohost_episode: a photo URL of the host instead (upload_file for a local file); a real person's face needs a paid plan
presenterNowith fromFilm: who presents and how. A face the video model flags stops it, nothing charged, with the choices to offer.
subtitlesNowhich on-screen text, once `captions` is on. LEAVE IT UNSET (or true) for SUBTITLES — every spoken word, in order, timed to the narration; no extra render, and there is NO cue limit, so the whole film is subtitled however long it runs (at most 4 words and one line a cue; 5 words / 32 characters in stills). Set it FALSE only if the user explicitly wants section HEADINGS instead: one short summary label held over each ~7-15s section. That is NOT what is being said — it is a label about it — so it is the wrong answer to "add captions" and to anyone watching on mute. `subtitles:true` also implies `captions:true`. TIMING: in the default blocks lane each cue is timed to its own spoken words (one listen to the narration, inside the quote; per-block timing, said, if it fails). In the stills lane cues are spread over each section’s measured narration by character count, exact at section edges, a few tenths off within one: it is not a word-level speech clock, so never promise frame-accurate sync there.
thumbnailNohost_episode: one thumbnail of the host (default true)
aspectRatioNo'9:16' default (a faceless channel video and a host episode default to 16:9)
frameDensityNohow many pictures per second of narration, and therefore what it costs. 'standard' (default) is a frame about every 1.5s — the density a stills film needs to read as a film rather than a slideshow; 'lean' is one about every 2.5s (the longest hold that still reads as a film, ~40% of the frames and ~40% of the cost); 'minimal' is one picture about every 3.5s, the cheapest and the longest any still is ever held, and it reads close to a slideshow. Only drop below the default if the user asked for something cheaper.
durationSecondsNoseconds (default 60): blocks lane 20-600, stills 20-120
idempotency_keyNochosen once, resent unchanged on retries: the same key never buys twice
expected_creditsNothe get_cost price; refused unbilled if the live price differs

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed6 schema fields changedv0.1.532
    • changedInput schema / properties / durationSeconds / description
      Previous value: -"target length 20-120s (default 60); drives the section count — ~10s of narration each, 3-8 sections"New value: +"seconds (default 60): blocks lane 20-600, stills 20-120"
    • addedInput schema / properties / expected_credits
      Added value: +{
      +  "description": "the get_cost price; refused unbilled if the live price differs",
      +  "type": "number"
      +}
    • addedInput schema / properties / get_cost
      Added value: +{
      +  "description": "true: return the price only; nothing runs or bills",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / idempotency_key
      Added value: +{
      +  "description": "chosen once, resent unchanged on retries: the same key never buys twice",
      +  "type": "string"
      +}
    • changedInput schema / properties / presenter / properties / voice / description
      Previous value: -"the presenter's voice (default: the film's own narrator)"New value: +"a voice to re-voice the presenter to, only when the user names one (default: the presenter's own voice from the video model)"
    • changedInput schema / properties / subtitles / description
      Previous value: -"which on-screen text, once `captions` is on. LEAVE IT UNSET (or true) for SUBTITLES — every spoken word, in order, timed to the narration; no extra render, and there is NO cue limit, so the whole film is subtitled however long it runs (at most 4 words and one line a cue; 5 words / 32 characters in stills). Set it FALSE only if the user explicitly wants section HEADINGS instead: one short summary label held over each ~7-15s section. That is NOT what is being said — it is a label about it — so it is the wrong answer to \"add captions\" and to anyone watching on mute. `subtitles:true` also implies `captions:true`. TIMING: in the default blocks lane each cue is timed to its own spoken words (one listen to the narration, a tiny charge inside the quote; per-block timing, said, if it fails). In the stills lane cues are spread over each section’s measured narration by character count, exact at section edges, a few tenths off within one: it is not a word-level speech clock, so never promise frame-accurate sync there."New value: +"which on-screen text, once `captions` is on. LEAVE IT UNSET (or true) for SUBTITLES — every spoken word, in order, timed to the narration; no extra render, and there is NO cue limit, so the whole film is subtitled however long it runs (at most 4 words and one line a cue; 5 words / 32 characters in stills). Set it FALSE only if the user explicitly wants section HEADINGS instead: one short summary label held over each ~7-15s section. That is NOT what is being said — it is a label about it — so it is the wrong answer to \"add captions\" and to anyone watching on mute. `subtitles:true` also implies `captions:true`. TIMING: in the default blocks lane each cue is timed to its own spoken words (one listen to the narration, inside the quote; per-block timing, said, if it fails). In the stills lane cues are spread over each section’s measured narration by character count, exact at section edges, a few tenths off within one: it is not a word-level speech clock, so never promise frame-accurate sync there."
  2. Changed5 schema fields changedv0.1.519
    • changedInput schema / properties / captions / description
      Previous value: -"turn ON-SCREEN TEXT on. DEFAULT FALSE, and leave it false unless the user asks — the narration already says the point and the pictures carry it, so the clean film is the better default. `captions:true` on its own burns SUBTITLES (see below), because that is what a caption is for: showing what is being said when the phone is on mute. Slim white CAPS, thin black outline, bottom safe band, no plate, no box."New value: +"turn ON-SCREEN TEXT on. DEFAULT FALSE, and leave it false unless the user asks — the narration already says the point and the pictures carry it, so the clean film is the better default. `captions:true` on its own burns SUBTITLES (see below), because that is what a caption is for: showing what is being said when the phone is on mute. Slim white text (CAPS in the stills lane), thin black outline, bottom safe band, no plate, no box."
    • changedInput schema / properties / format / description
      Previous value: -"blocks lane: 'explainer' (default; Gemini Omni 720p, 16:9 or 9:16, runs exactly the length asked) or 'faceless_channel' (a YouTube/TikTok faceless channel video; MiniMax H3 at 2K, 16:9 by default, five hard cuts per 10s block, a whole number of blocks). Omit and a history / kids / fairytale channel is a faceless channel video. 'host_episode' = the AI host episode (see the top)."New value: +"blocks lane: 'explainer' (default; Gemini Omni 720p, 16:9 or 9:16, runs exactly the length asked) or 'faceless_channel' (a YouTube/TikTok faceless channel video; MiniMax H3 at 2K, 16:9 by default, five hard cuts per 10s block, runs exactly the length asked: a last part-block is rendered whole and cut). Omit and a history / kids / fairytale channel is a faceless channel video. 'host_episode' = the AI host episode (see the top)."
    • addedInput schema / properties / fromFilm
      Added value: +{
      +  "description": "A PRESENTER ON A FINISHED FILM (only when the user asks for one): the job id of a finished video-blocks explainer or faceless-channel film. With `presenter`, that film comes back with a presenter speaking each block's line on screen; every block keeps its own picture and the film without the presenter is kept (cleanVideo). Never add a presenter unasked.",
      +  "type": "string"
      +}
    • addedInput schema / properties / presenter
      Added value: +{
      +  "description": "with fromFilm: who presents and how. A face the video model flags stops it, nothing charged, with the choices to offer.",
      +  "properties": {
      +    "creator": {
      +      "description": "a saved creator or a preset, by name or id (list_creators)",
      +      "type": "string"
      +    },
      +    "image": {
      +      "description": "or a photo / avatar of the presenter (upload_file for a local file); a real person's face follows the real-face rules",
      +      "type": "string"
      +    },
      +    "look": {
      +      "description": "'corner' (default): cut out of a green screen, standing in a bottom corner; 'badge': a round badge in a corner; 'split': the block on one half, the presenter on the other",
      +      "enum": [
      +        "corner",
      +        "badge",
      +        "split"
      +      ],
      +      "type": "string"
      +    },
      +    "position": {
      +      "description": "'br' (default), 'bl', 'tr' or 'tl' (a corner cutout stays at the bottom); for split, 'bl' puts the presenter first",
      +      "type": "string"
      +    },
      +    "scale": {
      +      "description": "the share of the frame the presenter takes, 0.2-0.8 (default: the look's own size)",
      +      "type": "number"
      +    },
      +    "voice": {
      +      "description": "the presenter's voice (default: the film's own narrator)",
      +      "type": "string"
      +    }
      +  },
      +  "type": "object"
      +}
    • changedInput schema / properties / subtitles / description
      Previous value: -"which on-screen text, once `captions` is on. LEAVE IT UNSET (or true) for SUBTITLES — every spoken word, in order, timed to the narration; free, no extra render, no extra credits, and there is NO cue limit, so the whole film is subtitled however long it runs (at most 5 words / 32 characters a line). Set it FALSE only if the user explicitly wants section HEADINGS instead: one short summary label held over each ~7-15s section. That is NOT what is being said — it is a label about it — so it is the wrong answer to \"add captions\" and to anyone watching on mute. `subtitles:true` also implies `captions:true`. TIMING: each cue is anchored to that section’s REAL measured narration length and distributed inside the section by character count — exact at every section boundary, approximate to a few tenths of a second within one. It is not a word-level speech clock, so never promise frame-accurate sync."New value: +"which on-screen text, once `captions` is on. LEAVE IT UNSET (or true) for SUBTITLES — every spoken word, in order, timed to the narration; no extra render, and there is NO cue limit, so the whole film is subtitled however long it runs (at most 4 words and one line a cue; 5 words / 32 characters in stills). Set it FALSE only if the user explicitly wants section HEADINGS instead: one short summary label held over each ~7-15s section. That is NOT what is being said — it is a label about it — so it is the wrong answer to \"add captions\" and to anyone watching on mute. `subtitles:true` also implies `captions:true`. TIMING: in the default blocks lane each cue is timed to its own spoken words (one listen to the narration, a tiny charge inside the quote; per-block timing, said, if it fails). In the stills lane cues are spread over each section’s measured narration by character count, exact at section edges, a few tenths off within one: it is not a word-level speech clock, so never promise frame-accurate sync there."
  3. Changed1 schema field changedv0.1.435
    • changedInput schema / properties / dryRun / description
      Previous value: -"true = return the exact credits this explainer reserves (its own pricing, stopped at the hold) and render nothing. Quote it before running one; try frameDensity lean or minimal when the balance is short."New value: +"true = return the estimated credits this explainer reserves (its own pricing, stopped at the hold) and render nothing. Quote it before running one; try frameDensity lean or minimal when the balance is short."
  4. Changed14 schema fields changedv0.1.366
    • changedInput schema / properties / aspectRatio / description
      Previous value: -"'9:16' default"New value: +"'9:16' default (a faceless channel video and a host episode default to 16:9)"
    • addedInput schema / properties / cameras
      Added value: +{
      +  "description": "host_episode: the rotation, ids frontal / three_quarter / close or framings in words; one entry = one fixed camera",
      +  "items": {
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • addedInput schema / properties / creator
      Added value: +{
      +  "description": "host_episode: the host, a saved creator or a preset by name or id (list_creators)",
      +  "type": "string"
      +}
    • addedInput schema / properties / dryRun
      Added value: +{
      +  "description": "true = return the exact credits this explainer reserves (its own pricing, stopped at the hold) and render nothing. Quote it before running one; try frameDensity lean or minimal when the balance is short.",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / format
      Added value: +{
      +  "description": "blocks lane: 'explainer' (default; Gemini Omni 720p, 16:9 or 9:16, runs exactly the length asked) or 'faceless_channel' (a YouTube/TikTok faceless channel video; MiniMax H3 at 2K, 16:9 by default, five hard cuts per 10s block, a whole number of blocks). Omit and a history / kids / fairytale channel is a faceless channel video. 'host_episode' = the AI host episode (see the top).",
      +  "enum": [
      +    "explainer",
      +    "faceless_channel",
      +    "host_episode"
      +  ],
      +  "type": "string"
      +}
    • changedInput schema / properties / frameDensity / description
      Previous value: -"how many pictures per second of narration, and therefore what it costs. 'standard' (default) is a frame about every 1.5s — the density a stills film needs to read as a film rather than a slideshow; 'lean' is one about every 2.5s (the longest hold that still reads as a film, ~40% of the frames and ~40% of the cost); 'minimal' is ONE picture per narration section, which is cheapest and is frankly a slideshow. Only drop below the default if the user asked for something cheaper."New value: +"how many pictures per second of narration, and therefore what it costs. 'standard' (default) is a frame about every 1.5s — the density a stills film needs to read as a film rather than a slideshow; 'lean' is one about every 2.5s (the longest hold that still reads as a film, ~40% of the frames and ~40% of the cost); 'minimal' is one picture about every 3.5s, the cheapest and the longest any still is ever held, and it reads close to a slideshow. Only drop below the default if the user asked for something cheaper."
    • addedInput schema / properties / fromDraft
      Added value: +{
      +  "description": "host_episode: a finished draft's job id, to render it in HD (720p) with the same script, cameras, set and host",
      +  "type": "string"
      +}
    • addedInput schema / properties / hostImage
      Added value: +{
      +  "description": "host_episode: a photo URL of the host instead (upload_file for a local file); a real person's face needs a paid plan",
      +  "type": "string"
      +}
    • addedInput schema / properties / lane
      Added value: +{
      +  "description": "'blocks' (default) = 10-second video blocks, real motion, one narrated line per block; 'stills' = the picture film (a still about every 1.5s, narrated, no video model; cheaper). Quote either with dryRun.",
      +  "enum": [
      +    "blocks",
      +    "stills"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / script
      Added value: +{
      +  "description": "host_episode: the exact words, said verbatim and split at natural breaks into 4-30s pieces",
      +  "type": "string"
      +}
    • addedInput schema / properties / setting
      Added value: +{
      +  "description": "host_episode: the set in words (default: written to fit the topic)",
      +  "type": "string"
      +}
    • addedInput schema / properties / thumbnail
      Added value: +{
      +  "description": "host_episode: one thumbnail of the host (default true)",
      +  "type": "boolean"
      +}
    • changedInput schema / properties / topic / description
      Previous value: -"what the explainer should teach or explain — a topic or a short brief"New value: +"what the explainer should teach or explain — a topic or a short brief (host_episode: or pass `script`)"
    • removedInput schema / required
      Removed value: -[
      -  "topic"
      -]
  5. Changed3 schema fields changedv0.1.285
    • changedInput schema / properties / music / description
      Previous value: -"music bed under the narration, measured to sit about 14 dB under the voice and sidechain-ducked beneath it. Omit and the KIDS and FAIRYTALE channels get their recommended bed COMPOSED for this film — those two are the only channels a bed is due on unasked, and it costs a small flat fee; every other channel ships dry. 'off' forces silence. 'library' takes a free curated track only, and ships dry when none is on file. NAME A MOOD — upbeat / calm / warm / epic / tense / playful / elegant / hype / chill / dramatic — to compose one on ANY channel, at the same fee. hermoso_capabilities reports the exact figure as explainerMusicCredits; quote it before you turn a bed on or pick a mood."New value: +"music bed under the narration, measured to sit about 14 dB under the voice and sidechain-ducked beneath it. Omit and the KIDS and FAIRYTALE channels get their recommended bed COMPOSED for this film — those two are the only channels a bed is due on unasked, and it costs a small flat fee; every other channel ships dry. 'off' forces silence. 'library' takes a free curated track only, and ships dry when none is on file. NAME A MOOD (upbeat / calm / warm / epic / tense / playful / elegant / hype / chill / dramatic) or DESCRIBE it in words to compose one on ANY channel, at the same fee. hermoso_capabilities reports the exact figure as explainerMusicCredits; quote it before you turn a bed on or pick a mood."
    • changedInput schema / properties / style / description
      Previous value: -"visual style. 'cinematic' (default) is photoreal; the rest are non-photoreal styled looks — editorial_collage (halftone cutouts + marker accents), flat_vector, stickman, whiteboard, ink_marker, silhouette, storybook (gouache), paper_diorama, isometric, claymation, pixel_art, watercolor, fluffy_toy (felted plush), low_poly, stylized_3d (matte clay render), studio_3d (preschool toy 3D on a white sweep — the Kids default), mannequin (clay-render reenactment figures — a History alternate). Ask the user which they want rather than picking silently; a styled pick costs more (see the cost note)."New value: +"visual style: 'cinematic' (default, photoreal); styled shortcuts editorial_collage, flat_vector, stickman, whiteboard, ink_marker, silhouette, storybook, paper_diorama, isometric, claymation, pixel_art, watercolor, fluffy_toy, low_poly, stylized_3d, studio_3d (the Kids default), mannequin; or ANY look described in words ('80s anime cel animation'), locked across every frame. Ask rather than pick silently; a styled look costs more."
    • removedInput schema / properties / style / enum
      Removed value: -[
      -  "cinematic",
      -  "editorial_collage",
      -  "flat_vector",
      -  "stickman",
      -  "whiteboard",
      -  "ink_marker",
      -  "silhouette",
      -  "storybook",
      -  "paper_diorama",
      -  "isometric",
      -  "claymation",
      -  "pixel_art",
      -  "watercolor",
      -  "fluffy_toy",
      -  "low_poly",
      -  "stylized_3d",
      -  "studio_3d",
      -  "mannequin"
      -]
  6. Changed1 schema field changedv0.1.225
    • changedInput schema / properties / endCard / description
      Previous value: -"append the branded end card (default true)"New value: +"append the branded end card. DEFAULT FALSE — set true ONLY when the user asks for one"
  7. Changed1 schema field changedv0.1.162
    • changedInput schema / properties / frameDensity / description
      Previous value: -"how many pictures per second of narration, and therefore what it costs. 'standard' (default) is a frame about every 1.5s — the density Higgsfield's own stills pipeline enforces; 'lean' is one about every 2.5s (the longest hold that still reads as a film, ~40% of the frames and ~40% of the cost); 'minimal' is ONE picture per narration section, which is cheapest and is frankly a slideshow. Only drop below the default if the user asked for something cheaper."New value: +"how many pictures per second of narration, and therefore what it costs. 'standard' (default) is a frame about every 1.5s — the density a stills film needs to read as a film rather than a slideshow; 'lean' is one about every 2.5s (the longest hold that still reads as a film, ~40% of the frames and ~40% of the cost); 'minimal' is ONE picture per narration section, which is cheapest and is frankly a slideshow. Only drop below the default if the user asked for something cheaper."
  8. Addedv0.1.161

TDQS

A4.7/5.0
Behavior5/5

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

Beyond annotations that mark it non-read-only, open-world, and non-destructive, the description discloses that it takes several minutes, returns a 480p draft, has a separate HD render path, and can incur significant render and upscale costs. It also covers defaults that materially affect behavior, such as captions being off by default and music being dry on most channels.

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?

It is dense and appropriately sized for a 28-parameter generative tool, with little wasted wording. However, the special host_episode case is front-loaded ahead of the general purpose, and the all-caps emphasis plus run-on clauses make it less scannable than it could be.

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 complex tool with no output schema, it covers the key operational facts: time, draft versus HD, lane behavior, captions/subtitles defaults, cost quoting, and where to find full rules. It does not describe the returned job/video retrieval shape, but that is a minor gap relative to the routing and cost information it provides.

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 100%, so the schema already carries the parameter-level detail. The description still adds useful cross-parameter meaning: the two lanes, host_episode's draft-to-HD flow, the captions/subtitles coupling, and when to quote dryRun costs.

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 description states a specific outcome: turn a topic into a finished narrated explainer video, with explicit distinctions for host_episode, lanes, and draft versus HD. It also names sibling tools that should be used instead for short product ads, so the agent can distinguish this tool from alternatives.

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?

It gives explicit routing: use render_ad, generate_video, or make_template_ad for a short product ad; use host_episode for a presenter episode; quote HD and upscale costs before running them. It also states when not to add captions, end cards, presenters, or upscales unasked.

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