Questions and answers

It gives your team one board showing everything your AI agents are waiting for: approvals, answers, reviews, access, and agents that have gone quiet. It tells the right person, and passes their answer back to the agent.

No. You add one MCP server and paste a short set of instructions into the file your agent already reads at the start of a session.

Any agent that can use an MCP server over HTTP with a bearer token, or call an HTTPS API. There are set-up steps for Claude Code and Codex.

Connecting an agent is three steps: register it, create a token, and paste the instructions into the file it reads. The getting-started guide shows each one. We have not yet timed this with enough teams to quote a number.

Yes. The board covers every project you can see and can be narrowed by project or by kind of wait. One agent can work on several projects, and each agent has its own tokens.

No. Holdups records the decision and makes it available, and the agent reads it the next time it checks in. If the agent’s session has ended, it has to be started again by you or your own tooling. Holdups does not do that today.

Two ways. The agent says so, by raising a blocker. And every agent that claims work sends a heartbeat: if the heartbeats stop without a handoff, the agent is flagged as stalled, after 15 minutes unless you change the limit. An agent that never reports at all is invisible to Holdups, which is why the instructions matter.

No. Issues stay in Jira and code review stays in GitHub. Holdups adds who each issue is waiting on, as a comment and three fields in Jira, and turns GitHub review requests into blockers that close when the review is done.

Jira Cloud and GitHub for the work, and email, Slack, Microsoft Teams and Telegram for notices. Messages in Slack, Teams and Telegram are one-way: people answer in Holdups, not in the chat.

The blocker closes, your answer is stored with it, and the agent reads it at its next check-in. If the issue is in Jira, it gets a comment and its waiting fields are cleared. You can also save the answer as a project decision, so no agent asks the same thing again.

Yes. Add them as a guest, and any blocker waiting on them arrives as an email with a link. The link works once, expires after 7 days, and lets them approve, decline or answer that one request and nothing else.

By email when something is assigned to them, in the project’s Slack, Teams or Telegram channel if one is set up, and in a morning digest. People can switch off their own emails. Admins always hear about blockers nobody owns and agents that have stalled.

Each kind of wait has a time limit, counted in the working hours of the person it is waiting on, so nothing escalates over a weekend. When the limit passes, the organisation’s admins are told. If someone is away, new blockers go to the stand-in they chose.

Yes. Each blocker keeps its conversation and a timeline of who did what. An audit log records every change and can be exported.

Holdups never connects to your repositories and never holds your AI provider keys. It stores what agents and people write in blockers, so tell your agents not to paste secrets into one. Sign-in supports two-factor, tracker tokens are encrypted, and data is hosted in the United Kingdom. It has had an internal security review, not an independent audit, and holds no certifications. Read the security page.

Everything on this site is built. It is in early access, by invitation and free of charge while that lasts. Pricing has not been set.

Fastidious LLP, part of Byte Digital Ltd, which has been building business software since 1998. Meet the team.