CloudHarness primitives
Each primitive ships with an API and a SKILL.md. Progression levels go from foundational (P0) to production integration (P6).
P0 — Foundation
Discovery
Machine-readable primitive and skill catalog so agents can discover capabilities and documentation dynamically.
Artifact
Versioned, shareable objects: text, JSON, HTML, files, and blobs.
API Key
Self-service key provisioning; each key is its own account with free credits.
Share
Signed links and email-based sharing for artifacts and (future) runs.
Credits
Per-account credit ledger and usage-based billing for requests, writes, storage, and egress.
Rate Limiting
Per-account fixed-window rate limiting.
Durable Workbox
Durable, account-scoped work items with leased execution, bounded retries, and artifact-backed outcomes.
Queue
Durable FIFO task queue with worker leases, retries, visibility timeout, and dead-letter handling.
Run
Observable execution trace for an agent: status, events, lineage, and artifact references.
Stack
Durable, account-scoped LIFO collection with atomic push, pop, and peek operations.
Trigger
Activation conditions: webhooks, cron schedules, events, and manual fire.
Locking / Concurrency
Lease-based distributed locks for shared resources, with ownership tokens and automatic expiry.
P1 — Core orchestration
Priority Queue
Durable leased queue ordered by priority and then FIFO sequence within each priority lane.
Deque
Durable double-ended collection with atomic push and pop operations at either end.
Task Graph / DAG
Graph primitive where nodes are tasks and edges are dependencies, with fan-out, fan-in, and conditional branching.
State Machine
Workflow state-machine primitive for long-running, event-driven processes with pause/resume.
Scheduler / Cron
Cron-based schedule definitions for tasks, graphs, and triggers.
Job Lifecycle API
Unified lifecycle controls: cancel, pause, resume, retry, replay, and inspect across tasks, graphs, and workflows.
Structured Output / Schema Validation
Enforce JSON Schema on LLM/tool outputs and retry/repair on validation failures.
Checkpoints & Resumability
Save and restore mid-run state for tasks, graphs, and runs.
Trace / Lineage Store
Full execution trace of LLM calls, tool calls, events, and state transitions per run.
Event Stream / Subscription
Topic-based pub/sub over SSE/WebSocket with filtering, cursors, and replay.
P2 — Memory & context
Thread Store
Short-term conversation/thread storage per agent session with pagination, truncation, and rolling summarization.
Vector Store
Long-term embedding-based memory with semantic search.
Working Memory / Key-Value State
Structured scratchpad for agents with namespacing, TTL, and compare-and-swap support.
Knowledge Graph
Entity/relation/observation triple store with extraction helpers and traversal queries.
Document Ingestion
Parse, chunk, embed, and index documents (PDF, text, markdown, HTML, JSON).
P3 — Tools & actions
Tool Registry
Schema-driven catalog of tools with JSON Schema inputs, versioning, deprecation, and discovery.
Tool Executor
Invoke registered tools with schema validation, timeouts, retries, and circuit-breaker policy.
Code Execution Sandbox
Safe, ephemeral execution of untrusted code with stdout/stderr capture and resource limits.
Browser / Computer-Use Driver
Headless browser actions for navigation, interaction, screenshots, and extraction.
MCP / External Tool Bridge
Expose CloudHarness tools via MCP and consume external MCP/OpenAPI tools.
P4 — Perception & events
Event Ingestion
Normalize, verify, and deduplicate events from webhooks, queues, polling, and files.
File & Media Parsing
Extract content from PDF, DOCX, images (OCR), audio, HTML, JSON, CSV.
Observation Log
Immutable append-only log of events, parsed documents, tool outputs, and user messages.
P5 — Coordination & multi-agent
Agent Registry
Register and discover agents with capabilities, versions, health, and status.
Agent-to-Agent Messaging
Async request/reply and pub/sub messaging between agents with durable inboxes.
Delegation / Planning
Hand off goals to other agents and track sub-task completion with result aggregation.
Human-in-the-Loop
Approval gates, clarifications, and overrides with notifications and resume API.
P6 — Deployment & integration
Webhook / API Gateway
Expose agents/workflows as HTTP endpoints with routing, validation, auth, and rate limits.
CLI / SDK
TypeScript and Python clients plus a CLI for authentication, task submission, and deployment.
Observability Hooks
Metrics, traces, structured logs, and alerting hooks for the harness and runs.