Skip to main content
Glama

Costory: Your Finops MCP

create_alert

Create a cost alert that monitors one or more queries and notifies when a condition fires. MCP is create-only — there is no update_alert; edit in the UI via the returned URL. Accepts the same query config as query (prefer datePreset over hand-computed from/to). The firing rule is a single condition boolean expression over the query names, e.g. a > 1000, rollingSum(a, 7, DAY) > 50000, or (a - timeShift(a, 1, DAY)) / timeShift(a, 1, DAY) > 0.2. Window math (rollingSum/weekToDateSum/monthToDateSum/timeShift) is evaluated daily in BigQuery, so you do NOT pick an evaluation period — instead set dedup to control re-notification frequency (CALENDAR once per WEEK/MONTH, or ROLLING once every N days). The period (datePreset or from/to) defines the preview/look-back window for the underlying queries. Use list_available_destinations for SLACK/TEAMS channel IDs. Returns a URL that you MUST include in your response so the user can view/edit the alert. EXAMPLE: "Alert me on Slack if our production AWS spend exceeds $50k over any 7 days, at most once a week" → { name: "Prod AWS weekly alert", queries: [{ type: "cost", name: "a", metricId: "cost", currency: "USD", filterCel: "cos_provider in ["AWS"] && cos_environment in ["prod"]" }], datePreset: "TRAILING_90_DAYS", condition: "rollingSum(a, 7, DAY) > 50000", dedup: { kind: "CALENDAR", calendarUnit: "WEEK" }, notificationChannel: "SLACK", slackChannelId: "C01ABC" }

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
toNoExplicit window end (inclusive, YYYY-MM-DD). Use with from instead of datePreset.
fromNoExplicit window start (YYYY-MM-DD). Use with to instead of datePreset.
nameYesDisplay name for the alert.
slugNoOrganization slug. Omit to auto-detect from your account (fails if you belong to multiple orgs).
aggByNoTime grain for the series: Hour, Day, Week, Month, or Period (default Month).Month
dedupYesDeduplication config controlling how often a still-firing alert notifies. Either CALENDAR (kind: CALENDAR, calendarUnit: WEEK | MONTH) = at most once per current ISO week / calendar month, or ROLLING (kind: ROLLING, windowDays: N) = at most once every N days.
limitNoMax groups/rows per series. Omit to use the default (100). Increase up to 1000 when you need more than 100 breakdown groups.
emailsNoEmail addresses (required if EMAIL)
compareNoAdd a comparison period to show cost evolution side-by-side. Omit `from`/`to` to compare against the preceding period automatically; set `chartType` to choose how it renders.
queriesNoSame series objects as the `query` tool `queries` array (cost / metric / usage / externalMetric / formula / budget). Each requires `type` (never omit) and a `name` (prefer short ids like a/b/c); put human labels in `alias`.
scopeIdNoOptional saved team scope id (from list_teams). Merges the scope whereClause into cost/usage queries.
conditionYesAlerts v3 firing rule: a single boolean expression over the query names (`name` field of each query). Supports arithmetic (+ - * /), comparisons (> >= < <= == !=), logical and/or/not, parentheses, and these window functions: rollingSum(a, N, UNIT) (trailing sum over the last N units, UNIT ∈ DAY|WEEK|MONTH, inclusive of today), weekToDateSum(a) (Monday-to-date), monthToDateSum(a) (1st-of-month-to-date), and timeShift(a, N, UNIT) (value shifted back N units; may wrap a window function). Examples: `a > 1000`, `rollingSum(a, 7, DAY) > 1000`, `(a - timeShift(a, 1, DAY)) / timeShift(a, 1, DAY) > 0.2`, `a > 10000 or rollingSum(a, 7, DAY) > 50000`.
datePresetNoOfficial date preset (same DatePreset as dashboards/reports, e.g. MTD, LAST_MONTH, TRAILING_30_DAYS). Prefer this over hand-computed from/to when a preset matches. Mutually exclusive with from/to.
slackChannelIdNoSlack target id (required if SLACK): a channel id (C…) to post to a channel, or a Slack user id (U…) to deliver a direct message to that user. Use list_available_destinations to discover both channels and the signed-in user's DM.
teamsChannelIdNoTeams channel ID (required if TEAMS)
notificationChannelYesNotification channel

TDQS

A5/5.0
Behavior5/5

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

The description discloses key behavioral traits beyond annotations: the tool is create-only (no update), window math is evaluated daily in BigQuery (so no evaluation period is chosen), dedup controls re-notification frequency, and the returned URL must be included in the response. Annotations only indicate create/write nature, so this added context is highly valuable.

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?

The description is detailed but every sentence earns its place. It is front-loaded with the core purpose, followed by critical limitations, parameter guidance, condition examples, and a complete example. No redundant or vague phrases; the density is justified by the tool's complexity.

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 (16 params, nested objects, no output schema), the description covers all critical aspects: condition syntax, dedup behavior, period semantics, notification channels, the return URL requirement, and a full example. It leaves no major ambiguity for correct invocation.

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 coverage is 100%, but the description adds significant meaning beyond the schema: it explains the relationship between period and evaluation, the condition expression syntax with examples, dedup semantics (CALENDAR vs ROLLING), and how to structure queries. The example maps a natural-language request to concrete parameter values, making parameter usage far clearer.

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+resource: 'Create a cost alert that monitors one or more queries and notifies when a condition fires.' It clearly distinguishes from sibling tools by noting 'MCP is create-only — there is no update_alert; edit in the UI via the returned URL.' This gives a precise scope and differentiates it from other create_* tools.

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?

The description provides explicit usage context and alternatives: 'MCP is create-only — there is no update_alert; edit in the UI via the returned URL', and 'Use list_available_destinations for SLACK/TEAMS channel IDs.' It also explains when to prefer datePreset over hand-computed from/to, and gives a full example. This goes well beyond basic when-to-use guidance.

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.

TDQS

A4.1/5.0
Disambiguation4/5

Tools are organized by resource (alerts, dashboards, reports, events, virtual dimensions) with distinct actions, so most are clearly separable. The main confusion risks are the three report-delivery side-effect tools (run_report_now, retry_report_execution, transfer_report_execution) and the generic get that spans five resource types, though detailed descriptions mitigate these.

Naming Consistency4/5

The dominant verb_noun pattern (create_*, list_*, update_*, preview_*, get_*) is consistent and predictable across the set. Deviations like bare verbs query/search/get and the noun-only virtual_dimension_overlap_matrix are readable but break the otherwise uniform convention.

Tool Count3/5

44 tools is heavy and exceeds the comfortable range, but the server covers a genuinely broad FinOps platform spanning querying, dashboards, reports, alerts, events, virtual dimensions, docs, skills, and suggestions. Each tool has a distinct job, though the sheer count makes agent navigation harder.

Completeness3/5

Core workflows are well covered: query → dashboard/report/alert/event, plus a full virtual-dimension draft lifecycle. Notable gaps include alerts being create-only with no update/delete, no deletes for dashboards/events/published virtual dimensions, and budget management limited to query/get with no create/update.

Resources