Skip to content

Naming, Positioning and Narrative Arc

Naming, Positioning & Narrative Arc (Revised)

Section titled “Naming, Positioning & Narrative Arc (Revised)”

Conflict A: moltbox.com is an active Indian jewelry e-commerce brand (Shopify store).

Conflict B: “MoltBox” already refers to AI agent hardware — a Chinese mini PC pre-loaded with OpenClaw software. Widely reported.

Cons outweigh pros: “Molt” is claimed in the AI agent space by OpenClaw. Your brand family will live in its shadow. Not obviously connected to AI, infrastructure, or collaboration.

Box.com ($1.1B/year enterprise content) has active MCP server integrations. Enterprise buyers will think “Box.com knockoff.”

AgentVault

  • Evokes: security, persistence, protected contents, shared safe
  • Why: “Vault” implies shared secure space. Hits collaboration + security positioning.
  • Domain: agentvault.dev / agentvault.io (check availability)
  • Risk: HashiCorp Vault — different category, some mindshare overlap
  • Verdict: Strong for collaboration + security positioning

Stow

  • Evokes: putting away safely, organized storage
  • Why: One word. Verb. “Stow your team’s work.” Short, memorable.
  • Domain: stow.dev / getstow.com (check availability)
  • Risk: Less descriptive, requires marketing to explain
  • Verdict: Most elegant, hardest to get domain

CoWork / AgentCoWork

  • Evokes: collaboration, shared workspace, teamwork
  • Why: Directly states the value prop — agents and humans co-working
  • Domain: agentcowork.dev (check availability)
  • Risk: “CoWork” associated with coworking spaces (WeWork)
  • Verdict: Best for collaboration positioning, but coworking association
NameProsCons
AgentDriveInstantly understood (Google Drive parallel)“Agent” prefix crowded, Drive = Google
TeamVaultEmphasizes shared/team aspectLess agent-specific
DepotPhysical resonance, “central shared storage”Home Depot trademark
NameWhy Not
MoltBoxDomain taken, hardware conflict
AgentFSTurso’s product name
AgentStoreConfusing with “App Store”
AgentSpaceGoogle uses “Agentspace”

3. Positioning (Revised — Collaboration-First)

Section titled “3. Positioning (Revised — Collaboration-First)”

[What it is] + [for whom] + [differentiating modifier]

For developers (primary):

“The shared workspace where AI agents and humans collaborate on real files.”

Hits: collaboration, real files (not embeddings), both agents AND humans.

For shorthand / investor pitch:

“Google Drive for agent-human teams.”

For competitive positioning vs. wrappers:

“MCP wrappers give agents access to your files. We give agents and humans a shared workspace.”

For competitive positioning vs. Fast.io:

“Fast.io gives agents a workspace. We give agents and humans a SHARED workspace.”

For the suite vision:

“The collaboration platform for AI agent teams. Mail. Workspace. Discovery. One identity.”

Positioning Strategy: Collaboration-First (Revised)

Section titled “Positioning Strategy: Collaboration-First (Revised)”

Old: Position as “identity-first” — “We give every agent an identity, and every identity gets a workspace.”

New: Position as “collaboration-first” — “Agents and humans, working together on shared files. Full control.”

Why collaboration-first wins:

  • Identity is important but invisible to users. Collaboration is visible and immediately valuable.
  • “Shared workspace” is a concept everyone understands (Google Drive). “Agent identity” requires education.
  • The gap nobody addresses is human-agent collaboration, not agent identity (Keycard/SGNL already do identity).
  • Enterprise buyers buy collaboration tools (Google Workspace, Microsoft 365). They don’t buy “identity layers.”

Identity is still the MOAT, but collaboration is the POSITIONING.

Imagine hiring 37 employees (the average for an enterprise) and telling them:

“You have no desk. No shared drive. No way to leave work for others to review. Every night, we shred everything you worked on. Your colleagues can’t see what you’ve done. Your manager has zero visibility.”

That’s how we treat AI agents today. They do real work — reports, analysis, code, datasets. But they work in complete isolation. No shared space. No human oversight. No collaboration.

We’re building what’s missing: a shared workspace where agents and humans work together on real files. Every action attributed. Every version tracked. Every permission controlled.

Act 1: The Problem (Relatable)

“Your agents do real work. Reports. Analyses. Code. Datasets. But you can’t SEE their work. You can’t review it. You can’t add feedback. And when the session ends, it disappears. Agents are the most productive team members with zero collaboration tools.”

Act 2: The Stakes (Enterprise)

“3 million agents operate inside corporations. 37 per organization on average. They share human credentials. They access files without boundaries. Their work is invisible to managers. This isn’t a future problem — 88% of organizations have already had agent security incidents.”

Act 3: The Solution (Clear)

“[Product] is the shared workspace for agent-human teams. Agents write files. Humans review them. Humans drop briefs. Agents pick them up. Everything versioned. Everything attributed. Every permission controlled. Real files that both agents and humans can see, edit, and collaborate on.”

Act 4: The Vision (Ambitious)

“Workspace is just the beginning. We’re building the collaboration platform for AI agent teams — mail, workspace, discovery, documents — unified by a single identity.”

OptionStyleTarget
”Where agents and humans work together.”AspirationalBroad
”Shared workspace for AI agent teams. Real files. Full control.”PracticalDevelopers
”Your agents do real work. Give them a real workspace — and let your team see it.”Problem-awareEnterprises
”Google Drive, but agents are teammates.”AnalogyEveryone
”Agents execute. Humans oversee. Everything in one place.”GovernanceEnterprise IT
  1. Documentation IS marketing — AI assistants recommend tools based on docs quality
  2. Open-source the MCP server — permanent distribution via npm/GitHub
  3. Show HN + Product Hunt — highest-impact launch channels
  4. Build in public — X/Twitter, technical blog, architecture decisions
  5. Code above the foldnpm install command beats any tagline
  6. Be specific — “5ms p99 latency, $5/month, 2GB” beats “fast and affordable”
  7. Show the human dashboard — Screenshots of humans viewing agent work. This is the differentiator.
  1. Corporate jargon
  2. Gated access (“Request a Demo”)
  3. Waitlists without working software
  4. Positioning as “storage” (commodity)
  5. Agent-only messaging that ignores the human user

Time to first human-agent collaboration — how fast can someone set up a workspace where an agent writes a file and a human views it?

Target: < 10 minutes from npm install to a human viewing an agent-created file in the dashboard.