How it works
- Personal. A key or sign-in acts as you, in one workspace only, and stops working if you leave the organisation.
- Review before Jira or ClickUp. Ideas sent from an AI tool arrive as drafts. Nothing is published until someone approves it in Pyloq.
- Confirmed changes. Moving a ticket only happens after you confirm it in your AI tool.
- Revocable. Disconnect any key or app instantly from Connectors → AI tools (MCP) in Pyloq.
- Visible. What AI tools do in Pyloq shows up in your workspace's activity feed.
See our Privacy Policy for how Pyloq handles your data.
Choose where Pyloq is used
Pyloq can't see which chat a request comes from, so you decide where it's used inside your AI tool. A connection always covers one workspace, in every chat where Pyloq is switched on.
- Per chat (claude.ai): in a chat, open the tools menu in the message box and switch Pyloq on or off.
- Per chat (ChatGPT): in a chat, choose which plugins it can use from the tools menu in the message box, and leave Pyloq off in chats that don't need it.
- Per tool (claude.ai): in Settings → Connectors → Pyloq, set each tool to always allow, ask first or never — for example, keep the read tools on and set
move_ticketandsubmit_requirementto ask first. - Per project (Claude Code):
claude mcp addsets Pyloq up for the folder you run it in; add--scope useronly if you want it in every project. - For your organisation: admins on Claude Team or Enterprise and ChatGPT Business or Enterprise can decide whether custom connectors like Pyloq are allowed, and for whom.
- Another workspace: disconnect Pyloq in your AI tool and connect again, picking the other workspace on the Pyloq sign-in page.
Menu names can differ slightly between plans and app versions.