cursor-sdk-automation
Skill0 starsChanged 6 months agoMITView on GitHub
Runs Cursor agents programmatically via @cursor/sdk or cursor-sdk for CI, bots, and backend automations. Use when integrating Cursor Agent API, Agent.create, streaming runs, or automating code tasks outside the IDE.
- Reads credentials
In this repository14 other files it gives its agents.
- a2a-agent-delegationSkill
- agent-memory-patternsSkill
- agent-observabilitySkill
- agent-redteam-securitySkill
- browser-agent-automationSkill
- build-mcp-serverSkill
- create-production-skillSkill
- deploy-agent-productionSkill
- evaluate-agent-qualitySkill
- github-ci-agentsSkill
- langgraph-multi-agentSkill
- multi-agent-supervisorSkill
- rag-agent-pipelineSkill
- secure-mcp-serverSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 24
Read the official SDK skill/docs in your environment before implementing — APIs evolve.
- Distinctive: line 52
Do not pass user PII in prompts to cloud runs
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Cursor SDK Automation
When to use
- GitHub Actions or cron jobs that need an agent to fix/review code
- Backend services that delegate coding subtasks to Cursor Cloud
- Scripts that resume long-running agent sessions
Packages
| Runtime | Package |
| TypeScript / Node | @cursor/sdk |
| Python | cursor-sdk |
Read the official SDK skill/docs in your environment before implementing — APIs evolve.
Workflow
- [ ] Store API key in CI secret (CURSOR_API_KEY), never in repo
- [ ] Create agent with repo + ref (branch/SHA)
- [ ] prompt() with clear, bounded task scope
- [ ] Stream or poll until terminal status
- [ ] Verify diff; run tests locally or in CI
- [ ] Cancel on timeout; handle rate limits with backoffTask design for automation
- One agent run = one PR-sized goal
- Include: repo path, files to touch, test command, definition of done
- Exclude: vague "improve everything" prompts
Error handling
- Retry transient 5xx with exponential backoff
- Surface
CursorAgentErrormessages to logs (redact keys) - Fail CI if agent stops without test pass
Security
- Scope API keys to minimum repos
- Do not pass user PII in prompts to cloud runs
- Review agent-produced diffs before merge (human or policy bot)
See github-ci-agents for Actions wiring.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.