Skip to main content
Glama

Triage

glitchless_triage
Read-onlyIdempotent

Ranked findings for a game, after its headline numbers (sessions, players, crash-free and error-free sessions, frame-time p50/p95/p99 and the share of frames over their own screen's budget, pct_frames_over_budget, with frame_budgets the share of play at each refresh rate; hitches a minute, rage quits) and the versions deployed in the window: performance spikes per level (with the gauges, counter bursts and events they correlate with), memory leaks, top errors, difficulty spikes / rage-quit hotspots, walls (a level, or one value of a level start's attribute such as a tier), version regressions, settings that run slow (a session property), the network (a whole lobby lagging at once: the server; one kind of connection, platform or region lagging; disconnects rising in a release) and the game's dedicated servers (falling behind when full, a crash that ends matches, memory that keeps climbing) — ranked by severity and player impact, each with evidence and a suggested fix, and naming a session property when most of a problem falls on one value. Findings the game's owner marked known (glitchless_mark_known) come after the rest with known and its note: accepted, not new. One that got worse since has came_back instead. Errors the owner resolved (glitchless_resolve_error) have no finding, unless one came back in a later version: then its finding has came_back with the version and the note. Start here when asked 'what's wrong with my game?'.

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": "glitchless_triage"
  }
}

Input Schema

NameRequiredDescriptionDefault
envNoOnly include sessions from this environment, e.g. "prod" (what the game passes to init; builds a developer runs usually send "dev")
gameNoThe game, by name or id, when signed in to a workspace (glitchless_list_games lists them); later calls keep it. With a key for one game (gl_sk_…), a call that names another game is refused: sign in to the workspace to reach every game.
hoursNoLook-back window in hours (default 24), as far back as the workspace keeps data on Glitchless Cloud (the answer says when it starts later than asked). Past 744 (31 days) answers are by day.
propsNoOnly sessions with these session properties (what the game set with setProperties): { rank: "diamond" }, or a list of values for any of them, { rank: ["diamond", "master"] }; "(none)" keeps the sessions without one. glitchless_events lists the properties and their values. Read from the sessions, so over 31 days or less
serviceNoGame service name (defaults to the most recently active game)
versionNoOnly include sessions from this game version
session_typeNo"user": players only; "tester": people on TestFlight and other test builds only; "synthetic": bots, playtests and store test devices only; "all" or unset: every session. Findings (glitchless_triage, and the overview's) read the production sessions of players unless env or session_type is given
steam_branchNoOnly include sessions that ran this Steam branch ("public", or a beta's name); games that use Steamworks send it

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / game / description
      Previous value: -"The game, by name or id, when signed in to a workspace (glitchless_list_games lists them); later calls keep it. Ignored with a key for one game."New value: +"The game, by name or id, when signed in to a workspace (glitchless_list_games lists them); later calls keep it. With a key for one game (gl_sk_…), a call that names another game is refused: sign in to the workspace to reach every game."
  2. First observed

TDQS

A4.1/5.0
Behavior5/5

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

Beyond the readOnly/idempotent annotations, the description discloses the output's ranking logic (severity and player impact), that each finding carries evidence and a suggested fix, and the ordering rules for owner-marked known findings (`known` + note, moved after the rest), regressions (`came_back`), and resolved errors (suppressed unless they recur, then `came_back` with version and note). It also notes that answers switch to daily granularity past 744 hours and that the response flags when data starts later than requested. That is substantial behavioral context the annotations and schema cannot supply.

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

Conciseness3/5

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

Every clause carries information, but it is delivered as one enormous run-on sentence with three levels of nested parentheses, so the scoring claim ('ranked by severity and player impact, each with evidence and a suggested fix') is buried mid-parenthesis rather than front-loaded. The single most actionable instruction, 'Start here when asked what's wrong with my game?', sits at the very end where an agent scanning for routing cues may miss it. A bulleted enumeration of finding families plus a leading one-line purpose statement would preserve the content at a fraction of the parsing cost.

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

Completeness4/5

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

With 8 parameters, all fully documented in the schema, and no output schema, the description correctly shoulders the burden of describing the return shape, which it does in unusual detail, including the state flags an agent must interpret. What is missing is boundary behavior: result caps, truncation, or what is returned when there is nothing to report, which matters for a scan tool aggregating a 9600-hour window.

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

Parameters4/5

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

Schema coverage is 100%, so the baseline is 3, but the description adds a genuinely useful cross-parameter rule: findings default to production player sessions unless env or session_type is given. That is default-selection semantics the schema only states locally per-property, and it clarifies how env and session_type interact for this specific tool.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens by naming the resource ('Ranked findings for a game') and then enumerates the finding families it produces (performance spikes, memory leaks, top errors, difficulty spikes, walls, version regressions, slow settings, network, dedicated servers), so the agent knows this is the catch-all ranked diagnosis tool rather than a single-metric query. It distinguishes itself from siblings like glitchless_performance or glitchless_network only indirectly, through scope breadth rather than an explicit contrast.

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?

The closing line 'Start here when asked "what's wrong with my game?"' is an explicit routing instruction and single-handedly establishes this as the entry point among 15 siblings. It also states the default population ('findings read the production sessions of players unless env or session_type is given'), which prevents a common misfire. It stops short of naming when NOT to use it or which sibling to pick for a known narrow question.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources