Skip to main content
Glama

Discover games (deals, new releases, Steam Deck, rating)

discover_games
Read-only

Find games across Steam by combining filters: discount, release date, platform, compatibility, review score, and tags. Each result includes price, reviews, compatibility statuses, and store link.

Instructions

Find games across the whole Steam catalog (keyless), filtered by ANY combination of: discount (min_discount — for 'what's on sale'), release recency (released_after / released_within_days — for 'new games'), hardware compatibility (steam_deck/steam_os/steam_machine/steam_frame — see each field's own description), native OS build (platform — windows/mac/linux), review quality (min_review / min_reviews), and user tags (tags — e.g. ['Roguelike', 'Deckbuilding'] for 'games like X'). Each result returns price/discount, review %, all four compat statuses, a vr_support flag (none/supported/required), popular tags, a clickable store_url, discount_end (when a deal expires) and release date in one call. Examples: '>80% off with 90%+ reviews' → set min_discount + min_review; 'recent well-reviewed games that run on Steam Deck' → set released_within_days + steam_deck + min_review; 'roguelike deckbuilders on sale' → tags:['Roguelike','Deckbuilding'] + min_discount. No appids needed — unlike get_items, which prices a list you already have. For 'games like X' from a SINGLE named title, get its tags via get_items and pass them here; for taste inferred from the player's WHOLE library instead, use get_recommended_games (key-gated). Note: min_discount is filtered server-side and re-checked client-side (so it holds at any value); setting released_after/released_within_days excludes not-yet-released games server-side, but the exact date cutoff — plus compat, platform, review and tag filtering — has no server-side support in the Steam catalog API, so those are scanned popularity-first and applied afterward over that same window — great for popular titles; a niche match may fall outside the top count (raise count for stricter filters). At most 60 results come back per call, best discount first: compare returned against matched to see whether the list was capped, and narrow the filters or page with start for the rest.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
tagsNoKeep only games carrying ALL of these user tags (case-insensitive), e.g. ['Roguelike','Deckbuilding']. Use exact Steam tag names — a misspelled/unrecognized one isn't an error, it just matches nothing. Applied over the scanned popularity window, so raise `count` when combining niche tags.
countNoHow many catalog entries to SCAN (1-200). Default 50. Not a result count — the filters below are applied over this window, so a strict combination can return far fewer than this; raise it for stricter filters. (Tools that cap what they RETURN call that `limit`.)
startNoPagination offset into the catalog (default 0).
countryNoCountry (cc) for prices/currency; overrides STEAM_COUNTRY for this call.
languageNoStore language (e.g. english, russian); overrides STEAM_LANGUAGE for this call.
platformNoNATIVE-build filter: keep only games shipping a native build for this OS (windows/mac/linux). 'linux' = a native Linux/SteamOS port. This is NOT Proton — for games that run via Proton compatibility use steam_os / steam_deck instead. Each result's `platforms` field lists its native builds, while steam_os/steam_deck report Proton compatibility, so native vs Proton stay distinct.
steam_osNoSteamOS compatibility — how well it runs on SteamOS in general (via Proton): 'verified' = SteamOS-Verified only; 'playable' = Playable or Verified. For a NATIVE Linux build instead, use platform:'linux'; for the Steam Machine console specifically, use steam_machine.
min_reviewNoMinimum positive-review %, e.g. 85. Applied over the returned page.
steam_deckNoSteam Deck compatibility (runs via Proton): 'verified' = Deck-Verified only; 'playable' = Playable or Verified. Not a native Linux build — see `platform` for that.
min_reviewsNoMinimum review count (filters out games with too few reviews).
steam_frameNoSteam Frame (VR headset) compatibility: 'verified' = Frame-Verified only; 'playable' = Playable or Verified.
min_discountNoMinimum discount %, e.g. 80 for '80%+ off' — this is the 'deals' filter. Omit to include full-price games.
steam_machineNoSteam Machine (Valve's console) compatibility (via Proton): 'verified' = Steam-Machine-Verified only; 'playable' = Playable or Verified. Its own rating, distinct from the general steam_os one.
released_afterNoKeep only games released on/after this date (YYYY-MM-DD).
released_within_daysNoAlternative to released_after: released within the last N days.

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
dealsYes
matchedYesHow many results survived every filter, out of the scanned window (see `count`).
returnedYesHow many of `matched` are in `deals` below. Lower than `matched` when the result was capped for response size — narrow the filters (or page with `start`) to see the rest.
total_matchingYesCount from whichever filters Steam actually applies server-side — min_discount, and (if released_after/released_within_days was set) excluding not-yet-released games — or the whole catalog's size if neither was given. NOT the number of games matching tags/platform/compat/review/the exact release-date cutoff, which have no server-side filter and this tool applies only over the scanned `count`-sized window below. Don't read this as 'N games match all my filters' — use `returned` for that instead.
Install Server

TDQS

A5/5.0
Behavior5/5

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

Discloses key behavioral traits: keyless operation, server-side vs client-side filtering, popularity-first scanning with potential niche misses, 60-result cap, and the need to compare returned vs matched. This goes well beyond the readOnlyHint/openWorldHint annotations, giving agents realistic expectations about result sparsity and pagination.

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?

Though long, the description is tightly structured: core function, examples, sibling distinction, then caveats. No redundant sentences; every clause carries operational meaning for a 15-parameter tool.

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 15-parameter discovery tool with output schema, the description covers filtering semantics, pagination, result cap, and alternative tools. It even addresses edge cases like misspelled tags and Proton-vs-native distinctions. Missing nothing critical for selecting and invoking the tool correctly.

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?

Schema already has 100% parameter coverage; the description adds cross-parameter meaning via examples (e.g., min_discount + min_review for '>80% off with 90%+ reviews'), explains that count is a scan window not a result limit, and clarifies server-side enforcement for min_discount and released_after. This is substantial added value beyond schema.

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?

Description opens with 'Find games across the whole Steam catalog (keyless)' – a specific verb + resource + scope. It enumerates all filter dimensions and explicitly contrasts itself with get_items and get_recommended_games, distinguishing it from siblings.

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?

Explicitly names alternatives: 'unlike get_items...' and 'use get_recommended_games' for library-based taste. Provides concrete use-case examples and clarifies when to use this tool vs others, including key-gating caveats.

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