Telexed

AI news through a solo-operator lens — only what changes your day3 of 829

FILTER[All][Agents & tools][Models & API][Generative media][Infra & SaaS][ASO & growth][Indie business][Idea signals][Other][★6+ high-signal]

Cursor Changelog ✕clear filters

Fri, May 291 dispatches

#0829
#0829Agents & tools Cursor Changelog2 days ago
`Cursor` Adds Auto-review Run Mode
90radar
CursorAI code editor — agent-driven code edits and review
Longer autonomous runs now need fewer approval prompts. The value is less babysitting, as long as the safety model is strong enough for real repo work.
- Auto-review is a new run mode for longer Cursor tasks with fewer approval prompts, reducing interruption during agent work.
- The changelog explicitly pairs fewer approvals with safer execution, so the feature targets autonomy without fully removing guardrails.
- Best fit is bounded repo work such as refactors, fixes, and test passes where repeated approvals slow the loop.
Source: cursor.com/changelog/auto-reviewRead original →
FIG-8291:1
90radar
FIG-8291:1

Wed, May 202 dispatches

#0828
#0828Agents & tools Cursor Changelog2 weeks ago
`Cursor` Adds Shared Canvases
70radar
CursorAI coding IDE — built-in agents and codebase context
Canvases can now be shared with a team. It turns planning or debugging boards into reusable collaboration artifacts, worth trying for async code work.
- Cursor canvases are now shareable with teammates; design notes and debugging context no longer need separate screenshots or docs.
- The update targets team collaboration, not model quality. Solo use stays similar unless you work with contractors or async reviewers.
- Reusable canvas context can reduce handoff friction when planning features, reviewing architecture, or tracking agent-assisted changes.
Source: cursor.com/changelog/shared-canvasesRead original →
FIG-8281:1
70radar
FIG-8281:1
#0827
#0827Agents & tools Cursor Changelog2 weeks ago
`Cursor Automations` Now Works Inside the Agents Window
80radar
CursorAI code editor — built-in agentic development workflows
Scheduled agent work is moving closer to the main coding surface. Multi-repo and repo-less runs make Cursor more useful for maintenance, audits, and non-code ops.
- Automations now appears in the Agents Window, reducing the gap between scheduled jobs and active agent work.
- A single automation can attach multiple repos, useful for cross-repo refactors, dependency checks, and shared package updates.
- Repo-less automations expand the use case beyond codebases: reminders, issue triage, release notes, or research queues can run without a project attached.
Source: cursor.com/changelog/05-20-26Read original →
FIG-8271:1
80radar
FIG-8271:1

`Cursor` Adds Auto-review Run Mode

`Cursor` Adds Shared Canvases

`Cursor Automations` Now Works Inside the Agents Window