Pushary sends your AI agents’ permission prompts and questions to the right person’s phone or Mac, and keeps a record of who said yes. Works with Claude Code, Codex, Cursor, Gemini CLI and Hermes, for one developer or a whole team.
This is the 6th launch from Pushary. View more
Pushary for Teams
Launching today
When an AI agent stops to ask permission, Pushary sends the question to the person running it, in their Mac’s notch or on their phone. Lee’s agent asks Lee. Sam’s agent asks Sam. A rule can send a push to your infra repo to your lead instead. Answers from the phone, Mac app and dashboard are saved with a name, so when something breaks, one page shows whose agent asked and who said yes. Works with Claude Code, Codex, Cursor, Agy and Hermes.
Free Options
Launch Team / Built With
Atlassian LoomHit record, hand off to AI
Promoted
Of all these notch related apps we've seen recently, this one looks like it might actually be useful. I'd need a longer trial to be sure though - something to consider.
@simon_moxon one month sound good? use PHTEAMS coupon in checkout.
I'm happy to guide you and your team through the full setup personally on a call https://cal.com/aadilghani/pushary
I love the permission monitoring capabilities this brings to larger teams and swarms of agents.
@pavel_hegler exactly! I think agents are really good are producing code and quality text, it's the human missing at that moment when agent needs a steer towards the right direction. Mix it with teams and you get an absolute unit or you get slopload.
Our bet and goal is to help teams steer their agents and still be accountable at the end of the day when discussing what went wrong and what can we do about it.
congrats on the launch aadil. for me the one command is anything that touches billing or drops a production table. one wrong yes and you spend the night restoring. does the phone prompt show the full command or only a short label?
@agshin_rajabov both phone and the notch shows the details with code diff and intent so you have an overview before approving, long story short you can sleep peacefully lol.
looking forward to welcoming you as our user.
Be awesome!
the "whose agent pushed this" question in slack is too real, the yes in someone's terminal just dissapears and nobody can find it after
the one i'd route to a second person is anything that emails customers from a prod account lol. can a rule match on what the command actually does, like a curl to a mail API??? or only on repo and path???
The audit trail is the part that sold me, "who said yes" is exactly what nobody can answer after an agent deletes something. One worry though: approving from a phone while walking is a one-tap habit. Is there any way to require a second person or a delay for the really destructive prompts, like prod deploys?
@galdayan Yes you can write rules which routes high stake actions to a lead for an approval.
Give it a shot and let us know how you like it.
Looking forward to welcoming you as our user!