Skip to main content
Glama

Harvest ~50 references per section (background)

inspo_harvest

Fills every app screen or website section with N design references from galleries and live sites, running in the background and streaming results to the board.

Instructions

Fill every section or app screen with N references (default 50). App screens (dashboard, tables, settings, login, onboarding, tab bar…) come from SaaS Interface, Mobbin (if logged in) and Dribbble filtered by platform. Website sections come from dedicated galleries (footer.design, navbar.gallery), Dribbble, and — the richest source — real award-winning sites (Awwwards, Siteinspire, plus sites you pick) automatically cut into sections: navbar, hero, about, catalog, footer… following links to /about, /shop, /pricing when a section lives on its own page. Runs in the background (several minutes) and streams into the board; returns immediately. Check with inspo_status.

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": "inspo_harvest"
  }
}

Input Schema

NameRequiredDescriptionDefault
queryNoIndustry keywords in English, 1-2 words (e.g. "restaurant", "coffee", "fintech"). Used on Awwwards, Siteinspire and Dribbble. With it, ~80% of each section comes from this industry and at most ~20% from general sources (more only if the industry runs out).
sitesNoLive sites you know are excellent for this niche/vibe. Crawled first.
targetNoReferences per section. Default 50.
sessionNoSession id. Defaults to the most recent session in this project.
maxSitesNoCap on live sites crawled. Default max(40, 1.5 × target).
platformNoPlatform for app screens. Default: mobile when the project type is mobile, else web.
sectionsNoSubset of the session sections. Default: all.
topicWordsNoWords that prove a live site is in the industry, in English and the project language, matched as prefixes against its title, description and headings (e.g. ["restaurant", "food", "menu", "cuisine", "chef", "dining", "comida", "restaurante"]). Default: the query words.
awwwardsCategoryNoAwwwards category slug, e.g. "food-drink", "fashion", "architecture", "e-commerce", "technology", "real-estate".

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.5.0

TDQS

A4.2/5.0
Behavior4/5

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

With no annotations, the description carries the burden and does so well: async execution, an estimated duration ("several minutes"), immediate return, streaming destination, and the follow-up status check are all disclosed, plus a conditional auth note for Mobbin ("if logged in") and crawl-depth behavior (following /about, /shop, /pricing links). It stops short of stating failure modes, whether repeated calls accumulate or replace existing references, or any rate/cost limits.

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?

Front-loaded with the core action and the default count, then organized by source domain (app screens vs website sections), with the operational caveats at the end. It is dense and long for a single paragraph — the source enumeration could be tightened — but no sentence is pure filler.

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

Completeness5/5

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

For a 9-parameter, no-output-schema, long-running tool, the description supplies exactly the missing pieces: async semantics, expected latency, where results land, and how to check progress, while the schema covers all parameter detail. Nothing essential to invoking it correctly is left out.

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 description coverage is 100% and the schema already explains query (including the ~80/20 industry mix), target defaults, maxSites, platform defaulting, and sections. The description adds only marginal meaning beyond that — it reinforces the default of 50 and the notion of "`sites` you pick" being crawled first. Baseline 3 applies when the schema does the heavy lifting.

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 ("Fill every section or app screen with N references") and immediately splits the domain into app screens versus website sections with named sources for each. It ends by routing the agent to the sibling monitoring tool ("Check with inspo_status"), so it is distinguishable from inspo_search/inspo_next_round without opening a schema.

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 clear operating context: this is a bulk background harvest that takes several minutes and streams to the board rather than returning data inline, and the agent should poll inspo_status afterward. It does not explicitly contrast with sibling entry points like inspo_search or inspo_next_round, so the when-not guidance is implied rather than stated.

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