SPB Git

spb/ultra-sharp-agent-skills Public

Ultra-Sharp Agent Skills — a research-first skill-authoring system + 72 production-ready skills for AI agents.

aiagentsskillsclaude
Python 100%
4.3 KB
View source Raw Blame History

# Examples — Executive Summary House Style

# Contents

  • Opening-sentence pairs
  • Complete worked example (bad → house style)
  • The "no recommendation yet" variant
  • Gotchas

# Opening-sentence pairs

❌ Never ✅ House style
"This report presents the findings of our infrastructure review." "Recommendation: consolidate to one cloud region, saving $84k/yr at negligible latency cost."
"Over the past quarter, the team has been analyzing churn." "Churn doubled in Q2 (4.1% → 8.3%); the driver is the March pricing change, and we recommend reverting it for accounts under 50 seats."
"Several options were considered for the mobile strategy." "We should ship the responsive web app now and defer the native app 12 months; this captures 80% of the value at 20% of the cost."

# Complete worked example

❌ Bad version (methodology-first, adjectives, no pick):

Executive Summary

As part of our ongoing commitment to operational excellence, a cross-functional working group was convened to evaluate our customer support tooling. The group conducted stakeholder interviews, reviewed industry benchmarks, and assessed three leading platforms across a comprehensive evaluation matrix. Each option presents meaningful benefits. Platform A offers robust automation. Platform B provides significant flexibility. Platform C is a strong contender on price. The working group presents these options for leadership consideration. Adoption of a new platform is expected to substantially improve response times and deliver considerable cost efficiencies.

✅ House-style version:

Executive Summary

Recommendation: adopt Platform A for customer support, replacing Zendesk in Q4. Net saving: $61k/yr; payback in 7 months (high confidence — pricing is contracted, migration quoted).

Why:

  • Automation resolves an estimated 38% of tier-1 tickets (measured in our 3-week pilot: 41% on 1,200 tickets).
  • First-response time drops from 6.2h to under 1h for automated categories.
  • Only option meeting our EU data-residency requirement out of the box.

Cost & risk:

  • $45k migration (one-time) + $9k/mo licence (vs $14k/mo today).
  • Risk: agent adoption → mitigation: 2-week parallel run, champions per team.
  • Risk: data-migration loss → mitigation: vendor-managed migration with row-count verification; Zendesk kept read-only for 90 days.

Next step: sign by Aug 15 (quote expiry); IT kickoff Aug 18. Owner: S. Boucher.

Methodology and full scoring matrix: appendix B.

# The "no recommendation yet" variant

Executive Summary

We cannot yet recommend a platform: the two finalists are tied on cost and features, and the deciding factor — EU data-residency certification for Platform B — is unconfirmed. Vendor answer due Aug 12; we will issue the recommendation Aug 13. If forced to decide today, Platform A (the reversible choice: month-to-month contract).

# Condensing an existing report (summary-at-the-top variant)

When the summary heads a longer document, it must stand alone — assume most readers stop after it.

❌ Never ✅ House style
"As detailed in section 4…" (forces the jump) State the number here; cite the section in parentheses for the curious.
"See below for our recommendation." The recommendation is sentence one of the summary itself.
Summarizing section by section in document order Reorganize into Recommendation → Rationale → Cost & Risk regardless of the report's structure.
New information that appears only in the summary Everything in the summary exists in the body; the summary selects, never adds.

# Gotchas

  • The summary written before the analysis ends becomes advocacy. Write the recommendation sentence only when you can defend its number.
  • "High/medium/low confidence" without basis is an adjective in disguise — attach what makes it high ("contracted pricing", "n=1,200 pilot").
  • Percentage without base ("38% of tickets") invites the first hostile question; give the base once ("of 1,200 pilot tickets").
  • Next step without an owner and date means no next step.
  • Executive summaries of bad news follow the same BLUF rule — burying "we lost the account" in paragraph three reads as spin, not tact.