agentsclimarketplace

Cursor sdk

Skill HKTITAN/cursor-sdk/skills/cursor-sdk

Entry point for the Cursor TypeScript SDK — install, auth, and the Agent.create / Agent.resume / Agent.prompt lifecycle. Use when starting any Cursor SDK integration or wiring durable agent IDs.From its SKILL.md

Install
npx -y skills add HKTITAN/cursor-sdk --skill cursor-sdk

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

3 things to look at

  • reads credentialsReads from 1 credential source: `CURSOR_API_KEY`.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
  • runs commandsInstructs the agent to run 1 command, including `npm install @cursor/sdk`.

SKILL.md

2.3 KB, 515 tokens by cl100k_base, as published. Nobody here has run it

cursor-sdk

Map of Content for the Cursor TypeScript SDK. Start here, then follow [[wikilinks]] to the atomic concept you need. Sibling skills (cursor-runs, cursor-cloud, cursor-mcp, cursor-subagents, cursor-models, cursor-hooks, cursor-errors) are part of the same graph — link across with [[../cursor-runs/references/stream]].

When to use

  • Bootstrapping any program that imports @cursor/sdk
  • Choosing between local execution, cloud execution, or pool/machine envs
  • Standing up durable agent IDs that survive process restarts
  • Resuming an existing agent rather than creating a new one

Install & auth

  • [[references/install]] — npm install @cursor/sdk, Node 20+, CURSOR_API_KEY
  • [[references/authentication]] — env var, explicit apiKey, service account keys, Team Admin caveat
  • [[references/account]] — Cursor.me(), Cursor.repositories.list()

Lifecycle

  • [[references/agent-create]] — Agent.create(options) full option surface
  • [[references/agent-resume]] — Agent.resume(agentId, partialOptions) and what is not persisted
  • [[references/agent-prompt]] — Agent.prompt(message, options) one-shot helper
  • [[references/agent-management]] — Agent.list, Agent.get, Agent.archive, Agent.unarchive, Agent.delete
  • [[references/disposal]] — await using, Symbol.asyncDispose, agent.close()

Local vs cloud

  • [[references/local-options]] — local.cwd, settingSources, sandboxOptions
  • [[../cursor-cloud/SKILL]] — cloud runtime, repos, PR auto, artifacts

Where to go next

  • Streaming and run control → [[../cursor-runs/SKILL]]
  • Adding tools via MCP → [[../cursor-mcp/SKILL]]
  • Picking and tuning models → [[../cursor-models/SKILL]]
  • Delegated subagents → [[../cursor-subagents/SKILL]]
  • Hook-based governance → [[../cursor-hooks/SKILL]]
  • Project rules and AGENTS.md → [[../cursor-rules/SKILL]]
  • Project skill packs → [[../cursor-skills-system/SKILL]]
  • Shell / CI alternative → [[../cursor-cli/SKILL]]
  • Error handling and retries → [[../cursor-errors/SKILL]]

What ships with it: 9 files

6.0 KB alongside SKILL.md

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.