Skip to main content
Glama
Sign In
Ferret

Ferret

by ferret-project
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Remote

Find the things in a room from photos

scan_room

Analyzes 1–5 room or open-box photos to list every item worth inventorying, with name, category and quantity. Runs in the background so you can bulk-add belongings after reviewing the results.

Instructions

Finds the things worth inventorying in 1–5 photos of a room or an open box: the vision model lists each with a name, a category and how many. Use it when the person wants to add a lot at once; for one thing use add_item_from_photo. Say where with locationId (a room) or containerItemId (a box). It looks in the background: about 20–60 seconds later room_scan_results has the list. Nothing is added until add_from_room_scan, after the person has said which to keep. One AI action per photo.

Input Schema

NameRequiredDescriptionDefault
photosYes1–5 photos of the room or open box, each base64
locationIdNoThe room they show, from list_locations
containerItemIdNoOr the box they show (an item with isContainer)

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv0.1.1
    • addedInput schema / properties / containerItemId / description
      Added value: +"Or the box they show (an item with isContainer)"
    • addedInput schema / properties / locationId / description
      Added value: +"The room they show, from list_locations"
    • addedInput schema / properties / photos / description
      Added value: +"1–5 photos of the room or open box, each base64"
  2. First observedv0.1.0

TDQS

A4.7/5.0
Behavior5/5

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

Annotations only say this is non-destructive and closed-world; the description adds what they cannot: it runs in the background (20–60 seconds), results land in room_scan_results, nothing is written until the person confirms via add_from_room_scan, and each photo costs one AI action. That staged-write semantics and latency/cost disclosure is exactly the context an agent needs before calling.

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

Conciseness5/5

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

Front-loads what it finds, then usage, then where to point it, then the async/confirmation contract — each sentence carries distinct operational information with no filler. Length is justified by the async workflow it must convey.

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 an async, side-effect-deferred AI tool with no output schema, the description covers timing, where results appear, what is not yet persisted, and the confirmation path. Nothing needed to call it correctly or to avoid a premature write is missing.

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% and the schema already documents photos as base64, the 1–5 bound, locationId from list_locations, and containerItemId as 'Or the box'. The description mostly restates that with 'Say where with locationId (a room) or containerItemId (a box)', adding framing but little new meaning, 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 ('Finds the things worth inventorying in 1–5 photos of a room or an open box') and even names the downstream output shape (name, category, count). It explicitly separates itself from the sibling add_item_from_photo, so an agent can choose without opening either schema.

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 the selecting condition ('when the person wants to add a lot at once') and the alternative for the opposite case ('for one thing use add_item_from_photo'). It also names the required follow-up (add_from_room_scan) and the results tool, so the whole workflow is routable from the description alone.

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