Skip to main content
Glama

decken_generate

Generate native editable PowerPoint (.pptx) decks from an outline, prompt, reference template, or rough slides. Rebuild existing decks or slide images into editable .pptx files.

Instructions

Primary Unified Slide Kernel (POST /v1/decks): (1) generate cohesive, expressive native .pptx directly from outline / prompt, (2) generate from outline + reference (.pptx template or reference image) inheriting palette & font DNA without rigid slot-stuffing, or (3) pass source (rough .pptx or slide image(s)) + optional prompt/reference to upgrade/reconstruct into 100% native editable .pptx.

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": {
      "outputPath": "<outputPath>"
    },
    "name": "decken_generate"
  }
}

Input Schema

NameRequiredDescriptionDefault
promptNoNatural-language brief or improvement instruction.
sourceNoLocal file path(s) or https:// URL(s) to an existing rough .pptx deck OR slide image(s) to improve/reconstruct.
modelIdNo
outlineNoMarkdown outline string (`## Slide 1 ...`) or structured slide array.
deckTitleNo
referenceNoLocal file path or https:// URL to a reference .pptx/.potx template OR reference slide image (.png/.jpg).
engineModeNovlm
outputPathYesDestination path to write the compiled .pptx file.
maxOutputTokensNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv1.1.0

TDQS

A3.5/5.0
Behavior3/5

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

With no annotations at all, the description carries the full behavioral burden. It usefully discloses that output is fully native/editable .pptx and that reference assets pass along palette and font DNA rather than being slot-stuffed, and `outputPath` implies a file write. But it is silent on the operational profile an agent most needs: whether this is an async job returned via `decken_job_status`/`decken_list_jobs`, expected duration, cost, failure modes, or permission requirements for `source` file access.

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?

It is a single front-loaded sentence that packs a lot of routing information efficiently, so nothing is truly wasted. But it is dense to the point of being hard to scan, mixes marketing phrasing ("cohesive, expressive," "100% native," "rigid slot-stuffing") with substance, and buries the three modes inside nested parentheticals rather than a scannable list.

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

Completeness3/5

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

For a 9-parameter generation tool with no output schema, the description should cover return shape and job semantics, and it does neither — no statement of what comes back (file path? job id? warnings?). It is complete on input composition, which is the complex part, but incomplete on the execution lifecycle.

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?

At 56% schema coverage, the description does real work for the four parameters that matter most: it explains that `source` accepts a rough deck or slide images, `reference` accepts a .pptx/.potx template or reference image, and `outline`/`prompt` drive direct generation. However, four parameters remain unexplained anywhere — notably `engineMode`, whose `vlm` vs `deterministic` enum is a consequential choice with no description in schema or prose, plus `modelId`, `deckTitle`, and `maxOutputTokens`.

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?

States a specific verb and resource (generate a native .pptx deck) and enumerates three distinct input modes keyed to `outline`/`prompt`, `reference`, and `source`. That is far more than a restatement of the name. It stops short of 5 because it never acknowledges the siblings `decken_reconstruct_deck` / `decken_reconstruct_slide`, whose names suggest overlap with mode (3), leaving an agent to guess which entry point wins.

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?

The three numbered modes effectively serve as when-to-use guidance by mapping intent to parameter combinations (text-to-deck, template/reference-driven, source reconstruction). There are no exclusions, however — nothing says when NOT to use this tool or when a sibling reconstruct/job tool is the better call, which is exactly the ambiguity this tool family creates.

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