agentleFS
OpenAIKnown publisheropenai/community-plugins/plugins/grafana-observability/skills/review-business-kpis/SKILL.md

review-business-kpis

Skill18 starsChanged 2 months agoApache-2.0View on GitHub

Review approved Grafana business-intelligence evidence for revenue, conversion, and operational KPIs. Use for bounded KPI scorecards, financial or operational metric checks, baseline comparisons, and business-unit reviews without changing source systems.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 3 are distinctive, and 8 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 8

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run the fixed grafana.business-kpis pack for an enrolled business unit or KPI set and keep financial and operational conclusions …

    line 8
  2. Very specific:

    Use only grafana_profile_status, grafana_list_packs, grafana_run_pack, grafana_get_evidence, and grafana_resolve_link.

    line 12
  3. Distinctive:

    Never accept or construct SQL, provider queries, URLs, headers, datasource IDs, panel JSON, variables, account identifiers, or raw database …

    line 14
  4. Distinctive:

    Keep Grafana, PostgreSQL and other datasources, dashboards, alerts, financial systems, and operational systems read-only.

    line 15
  5. Very specific:

    Call grafana_profile_status({profileAlias}).

    line 21
  6. Very specific:

    Call grafana_list_packs({profileAlias}) and require the exact active pack ID grafana.business-kpis.

    line 22
  7. Distinctive:

    Create one non-secret idempotency key and reuse it only for an exact retry.

    line 23
  8. Very specific:

    Call grafana_run_pack with that key, the aliases, packId: "grafana.business-kpis", origin, current window, and optional baseline window.

    line 24
  9. Very specific:

    Call grafana_get_evidence({reportRef,evidenceRef}) only for report-owned references needed to substantiate a material KPI claim.

    line 26
  10. Very specific:

    Call grafana_resolve_link({reportRef,linkRef}) only for a same-report link reference that helps a human inspect the enrolled Grafana view.

    line 27
  11. Very specific:

    Use regression language only when a returned issue has classification: SIGNED_RULE and kind: REGRESSION, supported by its qualifying …

    line 29

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Review Business KPIs

Run the fixed grafana.business-kpis pack for an enrolled business unit or KPI set and keep financial and operational conclusions evidence-bound.

Safety contract

  • Use only grafana_profile_status, grafana_list_packs, grafana_run_pack, grafana_get_evidence, and grafana_resolve_link.
  • Accept only profile and scope aliases; one declared LIVE, REPLAY, or SIMULATED origin; bounded current and baseline windows; and opaque references returned by these tools.
  • Never accept or construct SQL, provider queries, URLs, headers, datasource IDs, panel JSON, variables, account identifiers, or raw database requests.
  • Keep Grafana, PostgreSQL and other datasources, dashboards, alerts, financial systems, and operational systems read-only.
  • Confirm LIVE access. Do not mix LIVE, REPLAY, and SIMULATED values, and label simulated financial data prominently.

Workflow

  1. Establish profileAlias, an enrolled business or KPI scopeAlias, origin, and a current window no longer than seven days. Use an equal-duration, non-overlapping baseline for change claims.
  2. Call grafana_profile_status({profileAlias}). Stop on ERROR or blocked readiness; disclose PARTIAL readiness.
  3. Call grafana_list_packs({profileAlias}) and require the exact active pack ID grafana.business-kpis.
  4. Create one non-secret idempotency key and reuse it only for an exact retry.
  5. Call grafana_run_pack with that key, the aliases, packId: "grafana.business-kpis", origin, current window, and optional baseline window.
  6. Review revenue, conversion, and the enrolled operational KPI in returned order. Preserve returned units, severity, direction, nulls, and time basis; never add values with incompatible currencies or units.
  7. Call grafana_get_evidence({reportRef,evidenceRef}) only for report-owned references needed to substantiate a material KPI claim.
  8. Call grafana_resolve_link({reportRef,linkRef}) only for a same-report link reference that helps a human inspect the enrolled Grafana view. Never accept a URL as input.

Use regression language only when a returned issue has classification: SIGNED_RULE and kind: REGRESSION, supported by its qualifying baseline evidence. Treat classification: DEVIATION only as a deviation. Do not present observed KPIs as audited financial statements, forecasts, causal impact, or investment advice.

Interpret report status exactly

  • COMPLETE: expected KPI evidence completed; assess the returned signals and issues instead of assuming targets were met.
  • EMPTY: no usable KPI evidence was returned; do not substitute zeros.
  • STALE: evidence is too old for the requested reporting period; state the timestamps.
  • PARTIAL: a KPI, unit, source, or baseline is missing, failed, or mixed; limit comparisons to supported values.
  • CONTRADICTORY: KPI evidence conflicts; preserve both sides and do not choose a preferred business result.
  • ERROR: collection failed or was rejected; report the sanitized error and stop.

Completion format

Return scope, origin, reporting and baseline windows, overall status, a unit-aware KPI table, issues with their returned classification, gaps and contradictions, evidence references for every material claim, and only successfully resolved report-owned links.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.