HowLongToBeat MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| hltb_search_gamesA | Find games by title with estimated main story, extras and completion times. Returns only the first upstream page, not exhaustive results. No account or API key required. |
| hltb_get_gameA | Read a game detail page using an HLTB ID returned by search. Missing or unreadable pages produce an error, never fabricated durations. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 2 tools
The two tools have clearly distinct purposes: hltb_search_games finds games by title, while hltb_get_game retrieves a specific game's detail page by ID. The description explicitly links them (get uses an ID returned by search), so an agent cannot confuse the two.
Both tools follow a consistent hltb_ + verb_noun pattern (hltb_get_game, hltb_search_games) with the server prefix applied uniformly. The singular/plural difference is semantically natural, not a convention break.
Two tools is on the thin side, but it matches the narrow read-only scope of a game-length lookup service where search plus detail retrieval is the complete workflow. Nothing extraneous, though there is little room for agent flexibility.
The search-then-get pattern covers the core lookup lifecycle for this domain, and the descriptions transparently flag limits (first page only, error instead of fabricated data). No update/create/delete is expected for a read-only reference service, so there is no real dead end.