Skip to main content
Glama

Get personalized game recommendations

get_recommended_games
Read-only

Recommend unowned Steam games based on your most-played tags, weighted by review scores. Filter out genres or require a minimum discount for tailored suggestions.

Instructions

Recommend unowned Steam catalog games personalized to this player: tags on their own most-played owned games become weighted preferences (more playtime on a tag = more weight), discounted by each candidate's review score so a tag match on a poorly-received game doesn't outrank a better one, then ranked against a broad catalog page, excluding anything already owned. based_on_tags shows which of the player's own top tags drove the ranking; each result carries matched_tags and match_score alongside the usual price/review/compat card. Set exclude_tags to steer away from a genre despite it matching by playtime, and/or min_discount for a minimum deal size (e.g. 'suggest games on sale, not RPGs or shooters' → exclude_tags:['RPG','Shooter','FPS'], min_discount:30). Different from discover_games (which needs YOU to name the filters) — this infers taste from the player's WHOLE library instead, for 'what should I play next' / 'recommend me something'. For 'something like THIS ONE game' (a single named title), get its tags via get_items and call discover_games with them instead. Note: taste is weighted from only the player's 30 most-played owned games, and candidates come from a fixed 300-entry catalog scan, so a heavy exclude_tags/min_discount combination can return fewer than limit — there's no larger scan to fall back to. Requires STEAM_API_KEY and a public profile with game-details visible (same requirement as get_owned_games) — found:false is also returned if the player owns no games at all, or if too few of their played games have resolvable tags to build a taste profile.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
limitNoHow many recommendations to return (1-25). Default 10.
steamidNo17-digit SteamID64. Omit to use the STEAM_ID configured on the server. Convert a vanity/custom URL name with resolve_vanity_url first.
exclude_tagsNoDrop any candidate carrying ANY of these tags (case-insensitive), e.g. ['Souls-like'] for 'recommend me something except Souls-like'. Use exact Steam tag names — a misspelled/unrecognized one isn't an error, it just drops nothing.
min_discountNoMinimum discount %, e.g. 30 for '30%+ off'. Omit to include full-price games too. Filtered server-side and re-checked client-side, so it holds at any value including 100.

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Install Server

TDQS

A4.9/5.0
Behavior5/5

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

The description goes well beyond the readOnlyHint and openWorldHint annotations by disclosing the weighted-tag ranking logic, discounting by review score, exclusion of owned games, the 30-game/300-catalog scan limitations, and the edge case where found:false is returned. There is no contradiction with annotations.

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?

The description is detailed and every sentence contributes useful context, but it is longer than necessary. The opening sentence is dense with algorithm details, and some nuances (e.g., 're-checked client-side') could be trimmed for greater concision. Still well-structured and front-loaded.

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?

Given the tool's complexity, the description covers the algorithm, input/output behavior, failure modes, authentication requirements, and comparisons with alternatives. The presence of an output schema reduces the need to explain return values, and the description handles all other context comprehensively.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Although the schema already covers 100% of parameter descriptions, the tool description adds significant semantic value: it explains how exclude_tags interacts with playtime-derived preferences, gives a concrete example ('suggest games on sale, not RPGs or shooters' → exclude_tags:['RPG','Shooter','FPS'], min_discount:30), and clarifies that exclude_tags requires exact Steam tag names and that unrecognized tags silently drop nothing.

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 opens with a specific verb and resource: 'Recommend unowned Steam catalog games personalized to this player,' followed by a precise explanation of the recommendation algorithm. It clearly differentiates from sibling discover_games by stating that this tool infers taste from the whole library rather than requiring user-named filters.

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?

Explicit usage guidance is given: when to use this tool ('what should I play next' / 'recommend me something') and when not to (for a single named title, use discover_games after fetching tags via get_items). It also states prerequisites (STEAM_API_KEY, public profile with game-details visible) and notes that heavy exclude_tags/min_discount combinations can return fewer than limit.

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

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Grinv/steam-games-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server