Skip to main content
Glama

Make covers

make_cover

Kinetune’s cover art director makes 1 to 4 covers for a song, each a different idea The director reads the song (its meaning, never a lyric drawn literally), the user's choices and the artist's character, and the image model draws each cover; each is checked (no letters in the picture, reads as a thumbnail) and kept on the song (list_song_covers) as it is drawn, in about a minute. Only covers that pass are charged; the song keeps its cover until the user picks one (use_cover). style: how it is made (auto: the director picks); vibes: up to 2 moods (dark, sad, hood, luxury…); artist: auto (the director decides), always (needs their character), never; when the song is about two people (a love song, a duet) the director may show the other person beside the artist, partly or fully, always an invented person, never a real likeness (never with artist never); idea: the user's own words; words: none, title, or artist_title, lettered into the cover in the artist's typeface and read back letter by letter, always spelled as the song (the type style needs words); look: overrides the artist's colours and typeface. Ask the user for their choices rather than guessing, and quote first (quote_cover): credits must match. Needs a paid plan (402 plan_required) and an analysed song (409 song_not_ready); one round per song at a time (409 covers_in_progress).

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": {
      "song_id": "<song_id>",
      "credits": "<credits>"
    },
    "name": "make_cover"
  }
}

Input Schema

NameRequiredDescriptionDefault
ideaNoThe user's idea, in their words
lookNoThe artist's colours (hex) and typeface for this round
countNoHow many covers, each a different idea (1 to 4, default 1)
styleNoHow it is made (default auto: the director picks)
vibesNoUp to two moods or worlds
wordsNoWords on the cover, spelled exactly as the song: none (default), title, artist_title
artistNoWho is on it: auto (the director decides), always (the artist), never (no one). With the artist, a song about two people may also show the other person, always invented
creditsYescredits from quote_cover for this count, as shown to the user
song_idYesThe song id

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / artist / description
      Previous value: -"Who is on it: auto (the director decides), always (the artist), never"New value: +"Who is on it: auto (the director decides), always (the artist), never (no one). With the artist, a song about two people may also show the other person, always invented"
  2. Changed1 schema field changed
    • changedInput schema / properties / words / description
      Previous value: -"Words on the cover, set in real type: none (default), title, artist_title"New value: +"Words on the cover, spelled exactly as the song: none (default), title, artist_title"
  3. Added

TDQS

A4.4/5.0
Behavior4/5

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

Annotations only say this is a non-read-only, open-world, non-idempotent, non-destructive write. The description adds substantial behavior beyond that: roughly a minute per run, covers persist on the song as they are drawn, only covers that pass quality checks (no lettering, thumbnail-legible) are charged, and the three specific failure modes (402 plan_required, 409 song_not_ready, 409 covers_in_progress). It does not say what happens to credits on a failed round beyond the pass/fail rule, so not quite a 5.

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 content is high-value but delivered as one long run-on block chaining clauses with semicolons and colons, mixing purpose, parameter glosses, and error codes. It is front-loaded correctly, yet an agent must parse a wall of text to extract the parameter rules, which could have been separated into short labeled sentences.

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 9-parameter, nested-object generation tool with no output schema, the description covers cost, timing, persistence, prerequisites, and error handling well. The main gap is that it never states what the call returns (cover identifiers, a request id), leaving the agent to infer that from list_song_covers.

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 coverage is 100%, so the baseline is 3, but the prose adds genuine semantics: 'words' is lettered in the artist's typeface and read back letter by letter and must be spelled as the song; 'look' overrides the artist's colours and typeface; with two-person songs the second figure is always invented and never a real likeness (and never when artist=never). Some restatement of the schema exists (style auto, vibes up to two moods), keeping it below 5.

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?

Opens with a concrete verb and resource ('makes 1 to 4 covers for a song, each a different idea') and describes the pipeline: director reads the song, image model draws, each cover is checked. It explicitly names sibling tools (list_song_covers, use_cover, quote_cover) so an agent can place it in the workflow without opening other schemas.

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?

Gives explicit sequencing rules: quote first via quote_cover and make credits match, ask the user for choices rather than guessing, and use list_song_covers / use_cover for the follow-on steps. It also names the blocking preconditions (paid plan, analysed song, one round per song at a time) with their error codes, which is exactly the when-not-to-call guidance an agent needs.

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.

Resources