Skip to content

Kaze vs Paperclip — Complete Comparison

Part of Project Kaze ArchitectureDate: 2026-03-05


At a Glance

DimensionKazePaperclip
What it isAgent OS for SME clients (venture studio platform)Control plane for autonomous AI companies
StageArchitecture complete, core services implementedProduction-ready, full CRUD + governance
LanguageTypeScriptTypeScript
StackHono + Vercel AI SDK + Mem0 + pgvectorExpress 5 + Drizzle ORM + PostgreSQL (or PGlite)
Built bySpeedrun Ventures (venture studio)Paperclip AI (open-source)
LicensePrivateOpen-source
RepoMulti-repo (gateway, runtime, knowledge, agent-ops)Monorepo (pnpm workspaces)

1. Architecture Philosophy

AspectKazePaperclip
PatternDistributed services (gateway, runtime, knowledge)Monolith (Express API + React UI + PostgreSQL)
Agent modelVerticalAgent (long-running) + SubAgent (ephemeral per task)Heartbeat model (agents wake in short execution windows)
ExecutionContinuous — agents run within runtime, gateway holds LLM loopDiscrete — agents wake, checkout task, do work, sleep
Secret handlingGateway holds all secrets, runtime is credential-freeAgents receive env vars (API keys, endpoints) at invocation
CommunicationOpenClaw handles all user-facing interactionTask comments + dashboard UI (no chat)
DeploymentContainerized K8s + TailscaleDocker or local; embedded PGlite for zero-config

Key insight: Kaze is a distributed system where the gateway, runtime, and knowledge are separate services. Paperclip is a monolith where everything (API, UI, DB, agent invocation) lives in one process. Different trade-offs: Kaze has cleaner isolation but more operational complexity; Paperclip is simpler to deploy but harder to scale independently.


2. Agent Runtime

AspectKazePaperclip
Agent executionGateway runs LLM tool-use loop (Vercel AI SDK generateText + maxSteps)Agents are external processes — Paperclip invokes them, doesn't run LLM calls
Supported runtimesClaude/Gemini via gateway (single LLM provider)Bring-your-own: Claude Code, Codex, OpenClaw, shell, HTTP webhook
Tool systemBuilt-in ToolRegistry (github_api, file_glob, file_read, git_clone, etc.)No built-in domain tools — agents bring their own. Only Paperclip Skill for coordination
Skill definitionsYAML-based (prompt template + tool list per skill)Markdown-based (SOUL.md, HEARTBEAT.md injected into agent context)
Agent lifecycleLong-running VerticalAgent manages ephemeral SubAgentsHeartbeat: triggered by schedule, assignment, comment, or manual invocation

Key insight: Kaze is the agent runtime — it calls the LLM, executes tools, manages the loop. Paperclip orchestrates external agent runtimes — it doesn't care what the agent does internally, only that it follows the heartbeat protocol. This is a fundamental difference: Kaze has deeper control over execution, Paperclip has broader runtime compatibility.


3. Multi-Tenancy

AspectKazePaperclip
Isolation unitVertical + tenant pair (VerticalAgent per combination)Company (every entity scoped to company_id)
Data isolationPer-agent knowledge via Mem0 userId/agentId scopingFull company scoping — every table has company_id FK
Credential isolationGateway holds all secrets, scoped by configCompany secrets table (encrypted, versioned)
Org structureFlat — skills within verticalsHierarchical — agents have reports_to tree, company → goals → projects → tasks
Multi-companyDesigned for multi-tenant but not yet implemented at DB levelFirst-class — unlimited companies per instance, complete isolation

Key insight: Paperclip's multi-tenancy is production-complete (DB-level FK enforcement on every row). Kaze's is architected but relies on in-memory VerticalAgent scoping — the DB-level tenant isolation is not yet implemented.


4. Orchestration & Task Management

AspectKazePaperclip
Task dispatchOpenClaw → Runtime → SubAgent → Gateway → resultBoard/agent creates issue → agent checkout → heartbeat work → status update
Concurrency controlSingle SubAgent per task (ephemeral, dies after completion)Atomic checkout — 409 Conflict if two agents claim same task
Task hierarchyFlat (skill invocation, no parent-child tasks)Deep — company goal → project → parent issue → subtask chain
SchedulingNot yet implemented (planned: task scheduler)Built-in — cron-based heartbeat triggers, event-driven invocation
Run trackingObservation logger (planned)heartbeatRuns table — status, timing, events, error, external_run_id
Goal alignmentImplicit (skill prompt carries context)Explicit — every task traces ancestry back to company goal

Key insight: Paperclip has a complete project management system (goals, projects, issues, comments, labels, milestones). Kaze delegates task management to OpenClaw and focuses on skill execution. Paperclip's agents always know why they're doing something (goal ancestry); Kaze agents know what to do (skill prompt).


5. Supervision & Governance

AspectKazePaperclip
Approval modelPer-skill supervision ramp (supervised → sampling → autonomous)Per-action approval gates (hire agent, CEO strategy, extensible)
PromotionAutomatic — earned through track record (20/50 success thresholds)Manual — board approves/rejects explicitly
ScopeGranular per-skill (github-read vs github-write have different ramps)Coarse per-action-type (hiring, strategy — not per-tool)
DemotionAutomatic — 3 consecutive failures drops to supervisedManual — board can pause any agent at any time
AuditPlanned (observation logger)Complete — activityLog table, immutable, every mutation tracked
Config versioningNot implementedagentConfigRevisions — snapshot every config change with rollback

Key insight: Kaze's supervision ramp is more automated and granular (per-skill, auto-promote/demote). Paperclip's governance is more human-driven (board approvals, manual controls). Kaze optimizes for reducing human overhead over time; Paperclip optimizes for human oversight always being available.


6. Budget & Cost Control

AspectKazePaperclip
Cost trackingNot yet implemented (architecture references it)Production — costEvents table per token/cost event
GranularityPlanned: per-tenant, per-verticalPer-agent, per-company, per-project, per-goal, per-billing-code
Budget enforcementNot yet implementedHard-stop — agent auto-paused when spend >= budget
AlertingNot yet implemented80% utilization soft threshold
Monthly resetNot yet implementedAutomatic calendar-based UTC reset
Model selectionGateway modelHint (fast/balanced/best)Agent-level adapter config chooses model

Key insight: Paperclip has production-grade cost control that Kaze hasn't built yet. This is one of the clearest gaps — any multi-tenant agent platform needs per-tenant budget enforcement, and Paperclip's implementation is a strong reference.


7. Knowledge & Memory

AspectKazePaperclip
Memory systemDedicated service (Mem0 + pgvector + Google embeddings)None — no RAG, no vector search
Fact extractionAutomatic — Mem0 extracts facts from conversationsNone — context flows via task ancestry only
Semantic searchYes — per-agent vector similarity searchNone
Document ingestionBatch pipeline (docling + batch embedder)None
Context injectionOpenClaw before_agent_start hook queries knowledgeTask description + parent chain + workspace files

Key insight: Knowledge/memory is Kaze's strongest differentiator. Paperclip has zero built-in knowledge management — agents only know what's in their task description and ancestry. Kaze agents accumulate understanding over time via semantic memory.


8. Communication & UI

AspectKazePaperclip
User interactionOpenClaw (Slack, Telegram, WhatsApp, CLI, WebChat)React dashboard + CLI client
Agent communicationVia OpenClaw channels (conversational)Issue comments (task-oriented, threaded)
Real-timeOpenClaw WebSocket streamingWebSocket live events for dashboard
Agent-to-agentNot implemented (planned)Delegation via subtask creation + assignment
MCP supportConsidered (current discussion)Planned — 35 MCP operations defined for task management

Key insight: Kaze leans on OpenClaw for a conversational interface (chat-first). Paperclip is task-board-first (more like Linear/Jira for agents). Different UX paradigms — Kaze feels like talking to a colleague, Paperclip feels like managing a team.


9. Deployment & Operations

AspectKazePaperclip
Deployment4 separate services (gateway, runtime, knowledge, OpenClaw) on K8sSingle process (Express + embedded PGlite) or Docker
DatabasePostgreSQL + pgvector (external)PostgreSQL (external) or PGlite (embedded, zero-config)
Setup complexityHigh — multiple services, Tailscale, container orchestrationLow — npx paperclipai onboard --yes one-command setup
AuthNot yet implementedBetter Auth (sessions + API keys), board-claim flow
ModesProduction only (no local dev mode yet)local_trusted / authenticated+private / authenticated+public

10. Portability & Ecosystem

AspectKazePaperclip
Template systemYAML skill definitions per verticalCompany templates — export/import entire orgs
MarketplaceNoneClipHub (planned) — share company templates publicly
Agent portabilitySkills are YAML, but tied to Kaze's tool ecosystemFully portable — any runtime (Claude Code, Codex, OpenClaw, shell, HTTP)
Plugin systemNone (skills are the extensibility unit)Planned — embed custom plugins for reporting, knowledge

Summary: Where Each Excels

Paperclip is stronger at:

  • Governance — complete approval flows, audit trails, config versioning
  • Cost control — production-grade per-agent budget enforcement with auto-pause
  • Multi-tenancy — DB-level company scoping on every row
  • Deployment simplicity — single process, embedded DB, one-command setup
  • Runtime flexibility — any agent runtime works (Claude Code, Codex, OpenClaw, shell)
  • Task management — full project hierarchy with goal ancestry
  • Portability — export/import company templates

Kaze is stronger at:

  • Knowledge/memory — semantic search, fact extraction, document ingestion (Paperclip has none)
  • Supervision automation — per-skill ramps with auto-promote/demote
  • LLM control — direct tool-use loop management, model selection hints
  • Credential isolation — zero-secret runtime, gateway holds everything
  • Conversational UX — OpenClaw multi-channel chat vs task-board UI
  • Skill system — declarative YAML skills with prompt templating

Neither has yet:

  • Production MCP support (both planned)
  • Agent-to-agent communication (Paperclip has delegation via subtasks, Kaze has neither)
  • Self-improvement / meta-learning loops

Strategic Assessment

Paperclip and Kaze solve the same meta-problem (orchestrating AI agent workforces) but from opposite directions:

  • Paperclip is the org chart — it models companies, hierarchies, budgets, approvals. Agents are black boxes that follow a protocol.
  • Kaze is the agent brain — it models skills, knowledge, supervision ramps. The org structure is delegated to OpenClaw.

The most interesting question: should Kaze adopt Paperclip's governance/cost model? Paperclip's costEvents, activityLog, approvals, and company-scoped isolation are exactly what Kaze needs but hasn't built. Rather than reimplementing, Kaze could:

  1. Use Paperclip as the control plane — Paperclip manages companies, budgets, task dispatch. Kaze agents register as Paperclip adapters (like OpenClaw already can). Knowledge and supervision stay in Kaze.
  2. Port Paperclip's patterns — Copy the DB schema patterns (cost_events, activity_log, company scoping) into Kaze's services.
  3. Stay independent — Build governance from scratch within Kaze's architecture.

Option 1 is the most leverage for the least work, but adds a dependency. Option 2 is pragmatic. Option 3 is the most work for the most control.