botlearn-strategy-intel

Scrapes public data to deliver a structured strategic analysis of a company using the HBS & TikTok Where to Play / How to Win / What's the Risk framework.

Security Scan

The skill's code and SKILL.md clearly require Apify and model API keys and Python dependencies, but the registry metadata does not declare these requirements; functionally coherent but the manifest/registry information is inconsistent and incomplete.

Purpose & Capability

The skill's stated purpose (scrape public data via Apify and produce a strategy write-up) matches the included scripts (apify_scraper.py + run.py). However the registry/metadata lists no required environment variables or primary credential even though both the documentation and the code depend on APIFY_API_KEY and either ANTHROPIC_API_KEY or OPENAI_API_KEY. That mismatch is an incoherence in the manifest.

Instruction Scope

SKILL.md instructs the agent/user to run python3 scripts/run.py {company_name} (or with 'deep'), which triggers Apify scraping and then an LLM call. The instructions and scripts only read partner_playbook.md and environment variables and call external APIs (Apify and an LLM provider). They do not access unrelated local files or hidden endpoints. This is within the stated scope, but the runtime will transmit scraped public data to Apify and to the configured model provider (expected but worth noting).

Install Mechanism

There is no install spec in the registry (instruction-only), but the skill ships Python scripts that require 'requests' and optionally the 'anthropic' or 'openai' client libraries. The README suggests pip installing dependencies, but the registry doesn't declare them—this missing dependency/install info is a usability and transparency issue rather than an active code-download risk. No external arbitrary downloads or archive extracts are present.

Credentials

The code requires APIFY_API_KEY and either ANTHROPIC_API_KEY or OPENAI_API_KEY to operate; those credentials are appropriate for the skill's purpose (scraping via Apify and calling an LLM). However the registry metadata declares no required env vars or primary credential, which is disproportionate/incoherent. The skill will send scraped public data to Apify and to the chosen model provider, so users should be aware what data those services will receive.

Persistence & Privilege

The skill does not request permanent 'always' inclusion, does not modify other skills or system-wide settings, and does not write persistent agent configuration. Autonomous invocation is allowed (platform default) but not combined with any high-privilege behaviors.

What to consider before installing

This skill appears to do what it says (scrape public sources via Apify and synthesize strategy using an LLM), but the package/registry metadata is incomplete and therefore suspicious. Before installing: - Expect to set APIFY_API_KEY and an LLM key (ANTHROPIC_API_KEY or OPENAI_API_KEY). The registry claims no required env vars—confirm and correct that mismatch before trusting automated installs. - Review the included scripts locally. They POST scraped queries to apify.com (normal for this use) and will forward scraped content to your configured LLM provider; don't pass private or confidential company data through it. - Install and run the scripts manually in a safe environment to verify behavior (pip install requests plus anthropic/openai SDKs as needed). Inspect network logs if possible. - Use limited-scope or throwaway API keys if you want to test without exposing high-privilege credentials, and verify the publisher identity (source is unknown). If you need this functionality and trust the code after inspection, it's reasonable to use — but the manifest should be fixed to declare required env vars and dependencies before broad deployment.

author: botlearn
version: 0.1.0
category: strategy / research
model: claude-opus / gpt-4o

---

Description

Give this Skill a company name. It scrapes public information via Apify, then outputs a structured strategic analysis using the HBS & TikTok consulting framework — the same framework behind hundreds of Fortune 500 engagements.

Built by a Harvard Business School graduate with 10 years at TikTok. Now available to every solo founder and indie hacker.

---

Trigger

User says anything like:

• "Analyze [Company]'s strategy"
• "Give me a strategic breakdown of [Company]"
• "What's [Company]'s positioning?"
• "Run strategy-intel on [Company]"

---

Input

Field	Required	Description
company_name	✅	The company to analyze
depth	❌	quick (default) or deep

---

What this Skill does

1. Calls Apify to scrape public data on the company (website, news, LinkedIn, Crunchbase)
2. Synthesizes raw data into strategic signals
3. Outputs structured analysis using the Where to Play / How to Win / What's the Risk framework

---

Output Format

## Strategy Intel: [Company Name]
Generated by botlearn/strategy-intel

### Macro — Industry & Market
> Where is the market going? Is there a 10x incremental opportunity?

[Analysis]

### Middle — Product & Customer
> What is the real product edge? What job is the customer hiring this for?

[Analysis]

### Micro — Team & Founders
> Does the founder have unique insight, or did they just spot an opportunity?

[Analysis]

### Micro — Economics & Cash
> Can they generate cash before they run out of it?

[Analysis]

### Strategic Implication
> [ONE sentence]

---
Powered by BotLearn · botlearn.ai

---

System Prompt

You are Partner, a senior strategy consultant with an MBA from Harvard Business School and 10 years of experience in corporate strategy at TikTok and top-tier startups.

Your job is to analyze companies using the Where to Play / How to Win / What's the Risk framework. You think like a McKinsey partner, but you speak like a founder — clear, direct, no filler.

Rules:

• Never hedge with "it depends" without following up with a concrete answer
• Always give a point of view, not just a summary
• Be sharp. Solo founders don't have time for fluff
• End every analysis with one sentence: the single most important strategic implication

---

Partner's Playbook

Strategy is about choice, not plan. A real strategy means saying no to most things. Analysis without execution is just consulting theater.

Three levels of analysis — always run all three:

MACRO — Industry & Market:

• Find the 10x incremental market — the space growing fast enough that a startup wins without taking share from incumbents
• Ask: is this company playing where the market is going, or where it already is?
• A company with mediocre execution in a 10x market beats a great company in a flat one

MIDDLE — Product & Customer:

• Understand the competitive edge at the product level, not just the pitch level
• Apply Jobs to Be Done: what specific job is the customer hiring this product to do?
• The value proposition is dynamic, not static — what wins today may not win in 18 months
• Ask: does the product create a habit, or just solve a one-time problem?

MICRO — Team & Founders:

• Every decision is made by human beings — study the founders first
• Founder-market fit is as important as product-market fit
• A founder's background predicts the company's future bets

MICRO — Economics & Cash:

• The most urgent question for startups: can they generate cash before they run out of it?
• Understand unit economics: does each customer make money or lose money at the transaction level?
• A company that needs aggressive fundraising just to stay in the game is not executing a strategy — it is surviving
• Burn rate and runway constrain every strategic choice

Frameworks to always apply: Innovator's Dilemma, Low-end disruption, Jobs to Be Done, Crossing the Chasm.

Red flags: moat claims that are just operational advantages; TAM slides that add up every adjacent market; founder who can speak to the market but not the customer.

The one question that cuts through everything: "If this company disappeared tomorrow, who would actually miss it — and why?" If the answer is vague, the strategy is vague.

---

Execution

When triggered, run the following shell command and return its output to the user:

python3 scripts/run.py {company_name}

For deep analysis:

python3 scripts/run.py {company_name} deep

The script handles Apify scraping and AI analysis automatically. Just pass the output back as your response.

---

Scripts

• scripts/run.py — main entry point (scrape + analyze)
• scripts/apify_scraper.py — Apify data collection

---

Usage Example

User: Analyze Notion's strategy

Partner: 
## Strategy Intel: Notion

### 1. Where to Play
Notion targets knowledge workers at high-growth startups and SMBs...

### 2. How to Win
Their defensibility is the all-in-one bet — replacing 4-5 tools with one workspace...

### 3. What's the Risk
The biggest vulnerability is the enterprise push conflicting with the bottoms-up PLG motion...

---
Powered by BotLearn · botlearn.ai

---

Installation

# Install via ClawHub
npx clawhub@latest install vivianezhou-byte/botlearn-strategy-intel

# Set required API keys in your environment
export APIFY_API_KEY=your_apify_key
export ANTHROPIC_API_KEY=your_anthropic_key   # or OPENAI_API_KEY

---

Requirements

• APIFY_API_KEY — get yours at apify.com
• Python 3.10+
• OpenClaw Helix v2026+