Competitor Intel Stack — Sourced Battlecard & Freshness-Audit Claude Skill

Turn scattered competitor research into a dossier that states what it knows, what it doesn't, and how old each claim is — then a one-page battlecard sales and marketing can actually use in the moment, backed by a script that catches intel nobody has re-checked past its shelf life.

What this Claude Skill does

• Scopes the research to the actual decision it needs to inform, instead of open-ended "research this competitor" sprawl
• Forces every claim through a strict source-and-date format — claim — source: X, as-of: YYYY-MM-DD — with a clearly separated "unverified / inference" section for anything that can't be sourced, so nothing gets stated as fact without backup
• Scores each competitor on positioning clarity, pricing aggression, product momentum, and weakness surface, with every score tied to a cited evidence line — no unsupported scores allowed
• Compiles a one-page battlecard — say this, don't say this, watch for this, and the single most defensible wedge — the artifact your team actually reads in the moment, not the full dossier behind it
• Ships a working freshness-linter script that scans a compiled dossier and flags any claim past its staleness window, missing a source, or missing a date, so out-of-date intel doesn't quietly keep getting quoted in live calls

Built for

• Founders and sales teams who get "why not [competitor]" in calls and are currently answering from memory or a six-month-old doc
• Agencies and consultants who need a repeatable, defensible research format for client-facing competitive analysis
• Anyone who's had a rep confidently quote a competitor's old pricing in a live deal because nobody flagged that the intel had gone stale

What's inside

• SKILL.md — the full skill file with frontmatter and the five-phase methodology (scope, collect, score, synthesize, freshness gate)
• scripts/dossier_freshness_linter.py — a standard-library-only Python script that scans a compiled dossier and flags stale, unsourced, or undated claims, with a configurable staleness window
• templates/dossier-template.md and templates/battlecard-template.md — the structured outputs the skill compiles
• reference/source-checklist.md — six source categories to work through, plus the ethical boundaries on what's in and out of scope for collection
• reference/scoring-rubric.md — the four-axis scoring definitions with worked examples at each level
• LICENSE.md — usage terms

How a Claude Skill works

This is a Claude Skill — a single SKILL.md file Claude reads as instructions for a specific kind of work. Drop it in your Claude Project, .claude/skills/ folder, or wherever Claude can read it. Claude triggers the skill automatically when your prompt matches its description, then runs the methodology end to end.

Author

Built by Ukiyo Productions — a strategy-led branding and growth agency. Part of a 100+ skill marketplace built for founders, operators, consultants, and agencies.