docs: capture Paperclip positioning

docs/README.md

@@ -0,0 +1,7 @@
+# Docs
+
+Internal notes and working documents for AskClaw.
+
+## Strategy
+
+- [AskClaw vs Paperclip](strategy/askclaw-vs-paperclip.md)

docs/strategy/askclaw-vs-paperclip.md

@@ -0,0 +1,165 @@
+# AskClaw vs Paperclip
+
+This note captures the strategic split between AskClaw and [Paperclip](https://github.com/paperclipai/paperclip).
+
+## Short version
+
+Paperclip is open-source infrastructure for orchestrating **zero-human companies**.
+
+AskClaw should be the operating layer for **one-person companies**: one human founder, many agents, across the whole business.
+
+Cleaner:
+
+> Paperclip runs agent orgs. AskClaw helps a solo founder run a company with agents.
+
+Sharper:
+
+> Paperclip organizes agents into companies. AskClaw turns one founder into a company.
+
+## Core difference
+
+| Area | Paperclip | AskClaw |
+|---|---|---|
+| Core framing | Zero-human companies | One-person companies |
+| User role | Board/governor of autonomous agents | Founder/operator reviewing decisions |
+| Main abstraction | Agents, org charts, tasks, budgets, governance | Business outcomes, workflows, decisions, company execution |
+| Product feel | Agent ops dashboard / control plane | Founder-facing OPC command center |
+| Starting point | “Hire agents and assign goals” | “What should my company do next?” |
+| Scope | Strong on agent orchestration mechanics | Product, ops, support, growth, finance, docs, deployment |
+| Risk | Can become infra-heavy | Can become fluffy unless tied to real execution |
+| Best metaphor | Kubernetes/Jira for agent companies | Shopify/Notion/COO for a one-person company |
+
+## Strategic difference
+
+Paperclip says:
+
+> If OpenClaw is an employee, Paperclip is the company.
+
+AskClaw should not answer with:
+
+> No, we are also the company.
+
+That becomes mud wrestling.
+
+AskClaw should say:
+
+> You are the company. AskClaw gives every meaningful task an agent.
+
+This keeps the human founder central. That is the wedge.
+
+## Where Paperclip is strong
+
+Paperclip already claims the agent-infrastructure lane:
+
+- Bring-your-own agents: OpenClaw, Claude Code, Codex, Cursor, Bash, HTTP
+- Org charts
+- Budgets
+- Governance
+- Heartbeats
+- Tickets
+- Multi-company isolation
+- Company templates / Clipmart direction
+- Self-hosted open-source quickstart
+
+If AskClaw’s pitch is only “OPC control plane for agents,” Paperclip can punch it in the face. Politely. With a README.
+
+## Where AskClaw can win
+
+AskClaw should own the founder experience, not just the agent infrastructure.
+
+### 1. Decision layer
+
+AskClaw should answer:
+
+> Here are the three decisions that matter today.
+
+Not just:
+
+> Here are all your agent tasks.
+
+### 2. Business playbooks
+
+The product should ship opinionated playbooks for real solo-founder workflows:
+
+- Launch a product
+- Validate a market
+- Write docs
+- Handle users
+- Ship updates
+- Run a growth loop
+- Triage support
+- Operate infrastructure
+- Track cash, spend, and risk
+
+### 3. Taste and judgment
+
+AskClaw should not only dispatch agents. It should help decide what is worth doing.
+
+The founder should receive good next moves, not a bigger task queue.
+
+### 4. Cross-domain execution
+
+GitHub is only one surface. AskClaw should span:
+
+- Code
+- Content
+- Support
+- Analytics
+- Finance
+- Ops
+- CRM
+- Social
+- Deployment
+- Documentation
+
+### 5. One-human philosophy
+
+Paperclip is “zero-human company.”
+
+AskClaw should be “human amplified to company-scale.”
+
+## Positioning guidance
+
+Avoid public copy that sounds like Paperclip:
+
+- “AI agents run your company”
+- “Org chart for agents”
+- “Autonomous AI company control plane”
+- “Agent task board with budgets and governance”
+
+Prefer founder-centered copy:
+
+- “Turn one founder into a company.”
+- “Every meaningful task gets an agent.”
+- “Review decisions, not busywork.”
+- “Your company keeps moving even when you lose momentum.”
+- “Agents initiate, execute, escalate, and close real work.”
+
+## Best positioning line
+
+Broad audience:
+
+> AskClaw turns solo founders into one-person companies with AI agents.
+
+Technical/developer audience:
+
+> AskClaw is the OPC control layer for solo founders: every meaningful task gets an agent, and you review decisions instead of busywork.
+
+Founder-first wedge:
+
+> You are the company. AskClaw gives every meaningful task an agent.
+
+## Product implication
+
+AskClaw should not start by copying an agent org chart.
+
+It should start from founder outcomes:
+
+1. What matters today?
+2. What can agents do without me?
+3. What decisions need my judgment?
+4. What work is blocked?
+5. What changed while I was away?
+6. What should be shipped, answered, fixed, posted, or escalated next?
+
+The command center should make the founder feel less like a manager of bots and more like a company with leverage.