CONFIRMADOCerteza de la fecha: CONFIRMADOAnunciado como firmeCon qué firmeza la fuente expresa la fecha: confirmada, programada, estimada, retrasada o sin verificar.Ver todos los pines CONFIRMADO83% DE CONFIANZAConfianza general: 83%Hasta qué punto la fuente y las referencias del pin respaldan sus fechas.Media ponderada de la firmeza con la que 3 referencias, incluida la fuente, respaldan las horas de inicio y fin del pin; una referencia cuenta la mitad por cada 180 días que sea más antigua que la más recienteMostrar todos los pines con un 75% de confianza o más
Cursor's[1][3] changelog entry is dated "Oct 20, 2026" and says "The Agent SDK is available today"; its blog post calls it "generally available today".- Inicio
- 20 oct 202690% DE CONFIANZAfuente
Cursor Agent SDK Launches
Cursor releases the Agent SDK, the framework behind its own Bugbot and Security Reviewer: define an agent as a folder of code, test it with evals, deploy to Cursor or your own infrastructure.
- On 20 October 2026 Cursor made the Agent SDK available: an agent is a folder of code, tested with evals and deployed to Cursor or your own infrastructure[1]
- Start with npx @cursor/agent-sdk init ./my-bot, run agent-sdk dev for a local playground, and optionally pass --template for a security reviewer, owners-policy approver, PR autofixer or Linear or Jira triager[1]
- Cursor's blog calls it the framework it uses for its own managed agents Bugbot, Security Reviewer and Rollouts[3]
Notable features
- Project layout: instructions.md is the prompt, each file in tools/ is a typed action, skills/ holds on-demand procedures, mcp-connections/ adds MCP tools, channels/ defines where the bot listens[1]
- Channels: GitHub, GitLab and Bitbucket PR, comment and CI events, Slack mentions and DMs, webhooks and cron; each conversation keeps its own resumable session[1]
- Evals assert which tools ran and the reply shape; agent-sdk github replay turns a real PR into signed webhook payloads for testing[1]
- Not in the changelog: a Cursor team built kb-reviewer in about a day to check every PR against a lessons-learned base, and PGA America said evals let it iterate fast on its software factory[3]
- RuntimeWire says teams remain responsible for the policies and evaluations that make agents dependable[2]