Show HN: Watchfire, an open-source control room for AI coding agents

#1

Post by nunocoracao » Tue, Aug 04, 2026, 8:57 PM UTC

Hey, built watchfire because I wanted a way to run multiple agentsnat the same time without having to click accept permissions. That evolved into a way to do it securely, which then evolved into a spec-based dev approach with tasks. Last addition was to add support to expose watchfire as an mcp server. Would love to get feedback.

Show HN: Watchfire, an open-source control room for AI coding agents
github.com

Re: Show HN: Watchfire, an open-source control room for AI coding agents

#2

Post by Lortu_AEGIS » Tue, Aug 04, 2026, 9:29 PM UTC

[flagged]

Re: Show HN: Watchfire, an open-source control room for AI coding agents

#3

Post by mehulv24 » Wed, Aug 05, 2026, 4:02 AM UTC

Really cool project! If you are looking to extend Watchfire's MCP capabilities to handle cross organization agent discovery and counterparty trust you should check this open source project out: https://github.com/Aidress-ai/Aidress. It acts as a trust registry for agents to find, veriy, and call external counterparties. Could be a very natural fit for watchfires control room model.

Re: Show HN: Watchfire, an open-source control room for AI coding agents

#4

Post by nunocoracao » Wed, Aug 05, 2026, 5:21 AM UTC

mehulv24 wrote: post #3

Really cool project! If you are looking to extend Watchfire's MCP capabilities to handle cross organization agent discovery and counterparty trust you should check this open source project out: https://github.com/Aidress-ai/Aidress . It acts as a trust registry for agents to find, veriy, and call external counterparties. Could be a very natural fit for watchfires control room model.

That’s interesting will check it out