paperclip-ai-orchestration
Skill for using Paperclip — open-source orchestration platform for running autonomous AI-agent companies with org charts, budgets, governance, and heartbeats.
What this skill does
# Paperclip AI Orchestration
> Skill by [ara.so](https://ara.so) — Daily 2026 Skills collection.
Paperclip is an open-source Node.js + React platform that runs a **company made of AI agents**. It provides org charts, goal alignment, ticket-based task management, budget enforcement, heartbeat scheduling, governance, and a full audit log — so you manage business outcomes instead of individual agent sessions.
---
## Installation
### Quickstart (recommended)
```bash
npx paperclipai onboard --yes
```
This clones the repo, installs dependencies, seeds an embedded PostgreSQL database, and starts the server.
### Manual setup
```bash
git clone https://github.com/paperclipai/paperclip.git
cd paperclip
pnpm install
pnpm dev
```
**Requirements:**
- Node.js 20+
- pnpm 9.15+
The API server starts at `http://localhost:3100`. An embedded PostgreSQL database is created automatically — no manual DB setup needed for local development.
### Production setup
Point Paperclip at an external Postgres instance and object storage via environment variables:
```bash
# .env
DATABASE_URL=postgresql://user:password@host:5432/paperclip
STORAGE_BUCKET=your-s3-bucket
STORAGE_REGION=us-east-1
AWS_ACCESS_KEY_ID=$AWS_ACCESS_KEY_ID
AWS_SECRET_ACCESS_KEY=$AWS_SECRET_ACCESS_KEY
PORT=3100
```
---
## Key CLI Commands
```bash
pnpm dev # Start API + UI in development mode
pnpm build # Build for production
pnpm start # Start production server
pnpm db:migrate # Run pending database migrations
pnpm db:seed # Seed demo data
pnpm test # Run test suite
npx paperclipai onboard --yes # Full automated onboarding
```
---
## Core Concepts
| Concept | Description |
|---|---|
| **Company** | Top-level namespace. All agents, goals, tasks, and budgets are scoped to a company. |
| **Agent** | An AI worker (OpenClaw, Claude Code, Codex, Cursor, HTTP bot, Bash script). |
| **Goal** | Hierarchical business objective. Tasks inherit goal ancestry so agents know the "why". |
| **Task / Ticket** | A unit of work assigned to an agent. Conversations and tool calls are threaded to it. |
| **Heartbeat** | A cron-style schedule that wakes an agent to check for work or perform recurring tasks. |
| **Org Chart** | Hierarchical reporting structure. Agents have managers, direct reports, roles, and titles. |
| **Budget** | Monthly token/cost cap per agent. Atomic enforcement — agent stops when budget exhausted. |
| **Governance** | Approval gates for hires, strategy changes, and config rollbacks. You are the board. |
---
## REST API
The Paperclip API is served at `http://localhost:3100/api/v1`.
### Authentication
```typescript
// All requests require a bearer token
const headers = {
'Authorization': `Bearer ${process.env.PAPERCLIP_API_KEY}`,
'Content-Type': 'application/json',
};
```
### Create a Company
```typescript
const response = await fetch('http://localhost:3100/api/v1/companies', {
method: 'POST',
headers,
body: JSON.stringify({
name: 'NoteGenius Inc.',
mission: 'Build the #1 AI note-taking app to $1M MRR.',
slug: 'notegenius',
}),
});
const { company } = await response.json();
console.log(company.id); // "cmp_abc123"
```
### Register an Agent
```typescript
const agent = await fetch(`http://localhost:3100/api/v1/companies/${companyId}/agents`, {
method: 'POST',
headers,
body: JSON.stringify({
name: 'Alice',
role: 'CTO',
runtime: 'claude-code', // 'openclaw' | 'claude-code' | 'codex' | 'cursor' | 'bash' | 'http'
endpoint: process.env.ALICE_AGENT_ENDPOINT,
budget: {
monthly_usd: 200,
},
heartbeat: {
cron: '0 * * * *', // every hour
enabled: true,
},
reports_to: ceoAgentId, // parent in org chart
}),
}).then(r => r.json());
```
### Create a Goal
```typescript
const goal = await fetch(`http://localhost:3100/api/v1/companies/${companyId}/goals`, {
method: 'POST',
headers,
body: JSON.stringify({
title: 'Launch v1 to Product Hunt',
description: 'Ship the MVP and generate 500 upvotes on launch day.',
parent_goal_id: null, // null = top-level goal
owner_agent_id: ctoAgentId,
due_date: '2026-06-01',
}),
}).then(r => r.json());
```
### Create a Task
```typescript
const task = await fetch(`http://localhost:3100/api/v1/companies/${companyId}/tasks`, {
method: 'POST',
headers,
body: JSON.stringify({
title: 'Implement offline sync for notes',
description: 'Use CRDTs to merge note edits made offline. See ADR-004.',
assigned_to: engineerAgentId,
goal_id: goal.id, // links task to goal ancestry
priority: 'high',
}),
}).then(r => r.json());
console.log(task.id); // "tsk_xyz789"
```
### List Tasks for an Agent
```typescript
const { tasks } = await fetch(
`http://localhost:3100/api/v1/agents/${agentId}/tasks?status=open`,
{ headers }
).then(r => r.json());
```
### Post a Message to a Task Thread
```typescript
await fetch(`http://localhost:3100/api/v1/tasks/${taskId}/messages`, {
method: 'POST',
headers,
body: JSON.stringify({
role: 'agent',
content: 'Implemented CRDT merge logic. Tests passing. Ready for review.',
tool_calls: [
{
tool: 'bash',
input: 'pnpm test --filter=sync',
output: '42 tests passed in 3.1s',
},
],
}),
});
```
### Report Agent Cost
Agents self-report token usage; Paperclip enforces budget atomically:
```typescript
await fetch(`http://localhost:3100/api/v1/agents/${agentId}/cost`, {
method: 'POST',
headers,
body: JSON.stringify({
tokens_in: 12400,
tokens_out: 3800,
model: 'claude-opus-4-5',
task_id: taskId,
}),
});
```
### Heartbeat Ping
Agents call this endpoint on each scheduled wake-up:
```typescript
const { instructions, tasks } = await fetch(
`http://localhost:3100/api/v1/agents/${agentId}/heartbeat`,
{ method: 'POST', headers }
).then(r => r.json());
// instructions — what the org says to focus on now
// tasks — open tasks assigned to this agent
```
---
## TypeScript SDK Pattern
Wrap the REST API for cleaner agent integration:
```typescript
// lib/paperclip-client.ts
export class PaperclipClient {
private base: string;
private headers: Record<string, string>;
constructor(
base = process.env.PAPERCLIP_BASE_URL ?? 'http://localhost:3100',
apiKey = process.env.PAPERCLIP_API_KEY ?? '',
) {
this.base = `${base}/api/v1`;
this.headers = {
Authorization: `Bearer ${apiKey}`,
'Content-Type': 'application/json',
};
}
private async req<T>(path: string, init?: RequestInit): Promise<T> {
const res = await fetch(`${this.base}${path}`, {
...init,
headers: { ...this.headers, ...init?.headers },
});
if (!res.ok) {
const body = await res.text();
throw new Error(`Paperclip API ${res.status}: ${body}`);
}
return res.json() as Promise<T>;
}
heartbeat(agentId: string) {
return this.req<{ instructions: string; tasks: Task[] }>(
`/agents/${agentId}/heartbeat`,
{ method: 'POST' },
);
}
completeTask(taskId: string, summary: string) {
return this.req(`/tasks/${taskId}`, {
method: 'PATCH',
body: JSON.stringify({ status: 'done', completion_summary: summary }),
});
}
reportCost(agentId: string, payload: CostPayload) {
return this.req(`/agents/${agentId}/cost`, {
method: 'POST',
body: JSON.stringify(payload),
});
}
}
```
---
## Building an Agent That Works With Paperclip
A minimal agent loop that integrates with Paperclip:
```typescript
// agent.ts
import { PaperclipClient } from './lib/paperclip-client';
const client = new PaperclipClient();
const AGENT_ID = process.env.PAPERCLIP_AGENT_ID!;
async function runHeartbeat() {
console.log('[agent] heartbeat ping');
const { instructions, tasks } = await client.heartbeat(AGENT_ID);
for (const task of tasks) {
console.log(`[agent]Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.