BC competitor-research-playbook
Your competitor just launched. You have no idea how they grew so fast. Should you reverse-engineer their website? Track their social media? Map their growth flywheel? This gives you the complete SOP — from Wayback Machine snapshots to X/Twitter propagation analysis to growth flywheel scoring. Built from 150+ AI startup competitive analyses including the Lovable case study (4.3M views, 229K impressions launch day). By @WeiYipei. Inside: 4-step research framework (website → social → traffic → ad spend) · 3-version Wayback evolution analysis · X/Twitter propagation-chain mapping · growth-flywheel 6-stage scoring · KOL identification · content-effectiveness ranking · ICP + freemium pricing teardown · self-check checklist. New in 1.2: time-machine competitor archaeology — Google date-range search back to pre-fame years, star-history launch-day reconstruction (AppFlowy ~40-channel zero-budget launch teardown), competitor funding news as a launch-timing signal, and the "open source alternative" positioning play. Sourced from first-hand founder interviews (AFFiNE co-founder, 2026-04, self-reported figures marked). 🇨🇳 你的竞品刚刚爆了。你完全不知道他们怎么做到的。该拆官网？盯社媒？画增长飞轮？这份 SOP 给你从 Wayback Machine 快照到 X/Twitter 传播链路还原到飞轮六阶段评分的完整方法论。基于 150+ AI 创业公司竞品分析实战，含 Lovable 完整案例。 🇯🇵 競合が突然バズった。どうやって成長したか全く分からない。このSOPは、Waybackスナップショットからツイッター伝播チェーン分析、フライホイール6段階スコアリングまでの完全な競合調査フレームワークを提供します。150以上のAIスタートアップ分析から構築。 🇰🇷 경쟁사가 갑자기 폭발적으로 성장했습니다. 어떻게 그렇게 빨리 성장했는지 모릅니다. 이 SOP는 Wayback 스냅샷부터 X/Twitter 전파 체인 분석, 플라이휠 6단계 스코어링까지 완전한 경쟁사 조사 프레임워크를 제공합니다. 150개 이상의 AI 스타트업 분석에서 구축. Triggers: "competitor research" | "competitive analysis" | "competitor analysis" | "growth flywheel" | "market research" | "竞品调研" | "竞品分析" | "增长飞轮" | "传播链路" | "競合分析" | "경쟁사 분석" | "social media teardown" | "Twitter propagation" | "content strategy analysis"
As a process C 53/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- Shorten the description to 1024 characters.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 4. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1823 chars, limit 1024 - warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - note
description-budgetdescription takes 1823 of the ~15000-char shared budget for all skills - note
frontmatter-keyunknown frontmatter key "source"
Process rating: all ten parameters 53/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 100Tools and files. No external tools needed
- 100Steps. 22 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1394 tokens
- 100Running it twice. No mutating operations
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1822: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 10 example trigger phrases
- +4Structure: 31 headings
- +3Step-by-step instructions: 22 items
- +4Has examples (3 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 49.