All skills
deanpeters avatar

/battle-card-builder

@eb7a51e

Research and draft a competitive battle card from public evidence — every claim labeled and sourced. Use when a rep needs a field-action card, not a research report.

Use this Skill: https://skilld.dev/gh/deanpeters/product-manager-skills/battle-card-builder

This session only. Nothing lands on disk.

examplessample.md

≈1.5k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Battle Card Builder — Worked Example

All companies, products, URLs, and figures are fictional. This example completes the chain: the competitive-intel-watch example (run dated 2026-07-28) flagged the battle card's Pricing & Packaging and Do Not Say sections as stale after Wrenchline launched an entry tier. This card is the rebuild, using the snapshot and watch report already in session as its evidence base — the skill searched only for gaps and freshness, exactly as its Application step 1 instructs.

Invocation: Build a battle card: Fieldlight vs Wrenchline, shops of 10-50 techs, owner or office manager buying — triggered by the watch report flagging their new Essentials tier. Snapshot and watch report in session.


Battle Card: Fieldlight vs Wrenchline

As-of date: 2026-07-30 Deal context: trades shops of 10-50 techs; buyer is the owner or office manager

1. Thirty-Second Read

  • They win when: the shop already runs Wrenchline payroll or inventory — switching costs decide it — Fact (multi-year contracts, module lock-in per pricing page, Jul 2026)
  • We win when: tech adoption is the pain — their largest review complaint cluster is "my techs won't use it" — Inference (40+ reviews since Jan; review mining summary, Jul 2026)
  • The one thing to say: "Ask your techs to demo both apps for a day — we'll win that vote."

2. Say This

  • "Our tech app is the reason shops pick us — their own reviews say techs resist the Wrenchline app" — Inference (review cluster above; say "their reviews say," never more) — reviews, Jul 2026
  • "Essentials doesn't include the dispatch board — at 25+ techs you're forced into the full suite and its contract" — Fact (Essentials tier page, Jul 24, 2026)
  • "You can start on Fieldlight this week — no implementation project, no annual contract" — Fact (our pricing, Jul 2026)

3. Ask This (trap questions)

  • "What happens to your Essentials price when you cross 25 techs?" — documented answer: forced upgrade to the full suite, annual contract required — Fact (tier page fine print, Jul 24, 2026)
  • "Ask them how long a typical implementation runs" — documented answer: their own onboarding guide describes a 6-to-10-week implementation — Fact (onboarding guide, checked Jul 2026)

4. Watch Out For

  • Their depth pitch is real. Payroll, inventory, and fleet in one system — do not get drawn into a module-count comparison; we lose it — Fact (product pages, Jul 2026)
  • Likely counter: "Fieldlight is just scheduling — you'll outgrow it" — Inference (their "operating system" positioning implies it). Response is the roadmap conversation, not a feature claim we can't yet evidence.

5. Pricing & Packaging Snapshot

  • Essentials: $79/month under 25 techs (scheduling, quoting, invoicing; no dispatch board) — Fact (pricing page, as of Jul 24, 2026)
  • Full suite: per-tech pricing published to 100 techs; above that, "contact sales" hides the number — Fact (same page)
  • New since last card: Essentials launched Jul 2026 — treat any pre-July pricing talk track as dead

6. Do Not Say

  • Don't say "we're the affordable option" — Essentials at $79 contests it head-on; a rep leading with price now walks into their new tier — (watch report, Jul 28)
  • Don't say "Wrenchline can't do quoting for small shops" — Essentials includes quoting — Fact (tier page, Jul 24, 2026)
  • Don't claim their techs "hate" the app — the evidence is a review complaint cluster, an Inference; "their own reviews say techs resist it" is the defensible version

Appendix: Evidence Table

Claim Label Source Date
Essentials tier: $79/mo, <25 techs, no dispatch board Fact example.com/wrenchline/pricing 2026-07-24
Forced upgrade + annual contract at 25+ techs Fact example.com/wrenchline/pricing (fine print) 2026-07-24
Implementation runs 6-10 weeks Fact example.com/wrenchline/onboarding 2026-07
Tech-app resistance complaint cluster (40+ reviews) Inference example.com/reviews/wrenchline 2026-07
Multi-year contracts standard on full suite Fact example.com/wrenchline/pricing 2026-06

Assumptions to Validate

  • Essentials excludes the dispatch board permanently, not just at launch (carried from the watch report)
  • Our win rate actually improves in deals where a tech-app demo happens (instrument this)
  • The 6-10 week implementation figure holds for Essentials, not just the full suite

Why this example works

  • The chain fed the card. The competitor snapshots supplied the strengths and weaknesses; the watch report supplied the freshest layer (the Essentials tier) and named which sections to rebuild. The skill searched only for gaps — the onboarding guide was the one new fetch.
  • "Do Not Say" is doing its real job. The team's most beloved claim — "we're the affordable option" — died on July 24, and the card says so explicitly. That's the section protecting reps from their own folklore, four days after the folklore went stale.
  • Both trap questions cite documented answers. Each one's answer sits in the evidence table. The card contains no question a rep would be asking on hope.
  • Labels calibrate how hard a rep can lean. The tier facts are load-bearing (Fact, dated); the tech-adoption weakness is used, but phrased as "their reviews say" because it's an Inference — the card teaches the rep the defensible phrasing, not just the claim.

Source: SKILL.md on GitHub

1 warning2mo3 checks · Risk SAFE
  • Gen Agent Trust Hub2mo

    The 'battle-card-builder' skill is a workflow designed to research and draft competitive battle cards using public information. It instructs the agent to gather evidence from sources like pricing pages and reviews, then format the findings into a structured report with source citations and factual labels. No security risks or malicious patterns were detected.

  • Socket2mo

    No alerts

  • Snyk2mo

    Risk: MEDIUM · 1 issue

Signed by skilld at eb7a51e. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Activeupdated 3 months ago
argument-hint
[your product vs which competitor, and the deal context]
type
workflow
theme
market-intelligence
Other metadata
intent
Autonomous battle card construction: given your product and one competitor, the AI does the fieldwork — pricing pages, reviews, release notes, customer stories — and produces a field-action card with a source URL and Fact/Inference/Assumption label on every claim, sized to a rep's thirty seconds.
best_for
[
  "Arming sales against one competitor with claims they can actually defend",
  "Turning a competitive snapshot or watch report into a field-action artifact",
  "Building trap questions whose documented answers favor you"
]
scenarios
[
  "We keep hitting the same rival in enterprise deals — build the battle card from public evidence",
  "Our battle card is six months old and sales stopped trusting it; rebuild it with sources"
]
estimated_time
20-35 min per run

README badge

README badge for deanpeters/product-manager-skills/battle-card-builder