agentsclimarketplace

Board metrics operating review

Skill SylphxAI/skills/skills/board-metrics-operating-review

Public agent skills from SylphxAI — standards, product procedures, and one-command sync for Codex, Claude Code, and Grok Build

Install
npx -y skills add SylphxAI/skills --skill board-metrics-operating-review

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 19 days oldThe repository was created 19 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Prepare or audit the operating-metrics evidence packet behind a board or executive review: metric definitions, source reconciliation, period snapshots, cohort and segment cuts, variance bridges, forecast confidence, owner signoff, decisions, and action closure. Use when leadership needs board-ready operating truth. Do not use to invent company numbers, produce audited financial statements, design product analytics instrumentation, analyze one retention chart, or write investor persuasion detached from evidence.

SKILL.md

5.4 KB, ~1.0k tokens by cl100k_base, as published. Nobody here has run it

Board Metrics Operating Review

Turn source-backed operating metrics into decisions. Never turn missing data into plausible-looking board numbers.

Workflow

  1. Define the meeting, audience, reporting period, decision calendar, prior commitments, material risks, and board decisions required.
  2. Read references/board-metrics-operating-patterns.md.
  3. Establish current authority at use: metric dictionary/version, locked source extracts, finance close state, billing/CRM/product/support/incident sources, forecast version, currency/FX policy, and owner attestations.
  4. Build a metric-release ledger. For every requested metric, record definition, source, period, actual, plan, prior comparator, segment/cohort cut, caveat, confidence, owner, and release status.
  5. Reconcile boundary metrics before narrating them: accounting versus billing, bookings versus revenue, contracts versus subscriptions, product identity versus CRM account, gross margin cost allocation, and pipeline forecast.
  6. Explain material variance through a driver bridge and counterfactual: price, volume, mix, acquisition, activation, expansion, contraction, churn, cost, timing, one-offs, or data-quality change. Preserve arithmetic reconciliation.
  7. Link each narrative claim and board ask to released evidence. Present credible alternatives, recommendation, downside, owner, due date, and close condition.
  8. If values are absent or unlocked, emit not_provided/not_released cells, a source request register, symbolic formulas, and a blocked release decision. Do not generate example values unless the user explicitly supplies a sealed fictional dataset and asks for a fictional demonstration.
  9. Produce the released snapshot or blocked release packet, variance bridge, confidence ledger, board decision memo, and action register.

When not to use

  • For a single retention diagnosis, use retention-cohort-review.
  • For analytics event design or pipeline QA, use product-analytics-instrumentation-review.
  • For subscription pricing/packaging decisions, use saas-subscription-pricing; consume its approved decision rather than recreating it here.
  • For audited accounts, tax, securities disclosure, or legal assurance, prepare source mappings and open questions but defer authoritative conclusions.
  • For narrative-only rewriting, preserve the released facts and use the appropriate editing skill; do not let prose change metric meaning.

Source verification

  • Require exact locked extracts, metric and forecast versions, reporting periods, identity and currency rules, close state, derivation queries, and named owner signoff for every released value or claim.
  • Treat current finance, billing, CRM, analytics, support, and incident owners as separate authorities. Reconcile them explicitly; never use one dashboard or prior board deck as silent authority for another system.
  • Preserve supplied source bytes and calculation lineage. Missing, unlocked, stale, irreconcilable, or unsigned facts remain blocked cells rather than estimates generated for narrative completeness.

Guardrails

  • Never fabricate or interpolate a current-period value, benchmark, forecast, confidence, signoff, or variance driver.
  • Never release a derived metric whose definition, period, population, source, or reconciliation status is unresolved.
  • Never hide a missed plan behind blended growth, percentage-only reporting, changed denominators, or selective cohort windows.
  • Never present owner signoff as a sentence; retain who signed which version, when, against which caveat and source lock.
  • Keep financial/accounting classifications distinct from product-operating judgment, and identify where authorized finance review is required.
  • Never end with discussion themes only. Record decisions, dissent, actions, owners, due dates, required evidence, and close conditions.

Output

Board cycle and decision context:
- period / audience / prior commitments / decisions required / release status

Metric release ledger:
| Metric | Definition/version | Source lock | Actual | Plan | Prior | Variance | Confidence | Owner signoff | Release |
| --- | --- | --- | ---: | ---: | ---: | ---: | --- | --- | --- |
Use `not_provided` or `not_released:<reason>`; never substitute illustrative values.

Reconciliation and variance bridge:
| Bridge | Opening | Drivers | Closing | Reconciles | Evidence | Owner |
| --- | ---: | --- | ---: | --- | --- | --- |

Segment/cohort and forecast readback:
- cut / signal / caveat / decision implication

Board decisions and asks:
| Decision | Trigger evidence | Options | Recommendation | Downside | Owner | Due |
| --- | --- | --- | --- | --- | --- | --- |

Data requests and action register:
| Item | Source owner | Exact evidence/version | Due | Release effect | Close condition |
| --- | --- | --- | --- | --- | --- |

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.