quality-kpi-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| KPI_AS_OF | No | "Today" for all calculations, e.g. 2026-10-27 | latest date in the data |
| KPI_DATA_DIR | No | Folder with incidents.csv, deployments.csv, defects.csv | ./sample-data |
| KPI_OUTPUT_DIR | No | Where the dashboard is written | ./output |
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
} |
| prompts | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_data_sourcesA | What data is loaded: row counts, date range, services/components, and the as-of date used for all calculations. |
| get_mttrA | Incident response metrics for a period: mean time to restore (MTTR), acknowledge (MTTA) and detect (MTTD), with median, p90 and customer-detected share. Group by severity (default), service, month or root cause. |
| get_dora_metricsB | The four DORA keys (deployment frequency, lead time for changes, change failure rate, time to restore) with an Elite/High/Medium/Low tier for each and overall. |
| get_change_failure_breakdownB | Which services' deployments cause incidents or rollbacks, with rates, root causes and hotspots (services failing at 2x the overall rate). |
| get_defect_metricsB | Defects created and resolved, escape rate (share found in production), reopen rate, escape rate by component, and resolution time and SLA compliance by priority. |
| get_defect_agingA | Open defects right now: age buckets, open count by priority, and every defect past its SLA (Critical 2d, High 7d, Medium 30d, Low 90d), worst first. |
| get_kpi_trendB | Month-by-month series for one KPI. Metrics: mttr_hours, mtta_minutes, mttd_minutes, incidents, customer_detected_pct, deploys_per_week_per_service, lead_time_hours, change_failure_rate_pct, defects_created, escape_rate_pct, open_defects_end_of_month. |
| compare_periodsA | Every KPI side by side for two periods with % change and improved/worse/flat. Defaults to the chosen period vs. the equal-length period before it, e.g. this quarter vs. last. |
| generate_executive_summaryA | A one-page markdown quality report for leadership: bottom line, KPI scorecard vs. the previous period, DORA tiers, wins, risks, and recommended asks. Ready for email, Slack or Confluence. |
| export_leadership_dashboardA | Write a self-contained HTML dashboard (KPI tiles, wins and risks, MTTR / change failure / defect trends, DORA tiers, SLA breaches) to the output folder and return its path. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| monthly_quality_review | Prepare the monthly engineering quality update for leadership. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 10 tools
Each tool has a clearly distinct purpose: data inventory, defect aging, defect metrics, incident MTTR, DORA metrics, change-failure breakdown, KPI trends, period comparison, executive summary, and dashboard export. Overlaps such as time-to-restore appearing in both MTTR and DORA tools are framed at different levels of aggregation, and the descriptions make the boundaries clear.
All tools use consistent snake_case with a clear verb_noun pattern: get_, list_, compare_, generate_, and export_. The naming is predictable and readable throughout, including acronym cases like get_mttr and get_dora_metrics.
Ten tools is well-scoped for a quality KPI reporting server, with separate tools for each major metric family, reporting workflow, and data discovery need. No tool appears redundant or purely decorative.
The surface covers the domain end to end: discovering loaded data, retrieving defect, incident, DORA, and change-failure metrics, trending and comparing KPIs, and producing executive-level summaries and dashboards. No obvious operational gap would block the intended reporting workflows.