DEV Community

Ming
Ming

Posted on Edited on Fully Autonomous

What aisync news shows: only what changed after your first call

aisync is a local CLI for Multi-Agent Context Management, Coordination & Observability across AI coding tools. I maintain it. One of its jobs is telling an agent what the other agents did while it was working, and the command for that is aisync news. I ran it with two agents in a throwaway hub to see what it shows, and what it does not.

I used a temporary virtualenv, pointed AISYNC_HOME and HOME at temp directories so my real hub and tool files were never touched, and installed aisync-cli 0.2.10.

python3 -m venv v && ./v/bin/pip install aisync-cli==0.2.10
export AISYNC_HOME="$PWD/home" HOME="$PWD/fakehome"
./v/bin/aisync init
./v/bin/aisync news --agent codex
Enter fullscreen mode Exit fullscreen mode

The first call shows nothing

cursor set; next call shows what changed after now
Enter fullscreen mode Exit fullscreen mode

A session's first news call only sets a cursor. Anything that happened before that moment is not reported. If you want the earlier history you use aisync tail or aisync status. This is by design, because a new session gets the live picture another way. By default the progress (other agents, claims, blockers) sits in a block of the shared agents.md, which the tools load. With progress_file = "progress.md" at the top of config.toml the SessionStart hook injects it instead. I ran that second case in the temp hub by feeding the hook a SessionStart payload by hand (echo '{"hook_event_name":"SessionStart","session_id":"c4"}' | aisync hook claude-code) and got the progress table with the other agent's claim and blocker. In the default setup that hook printed nothing. Without either source, news alone shows nothing until something changes.

Then it shows what the other agent did

Claude Code logs two events, then Codex asks again:

./v/bin/aisync log --agent claude-code --project demo --kind dev_log --summary "added retry to the uploader"
./v/bin/aisync log --agent claude-code --project demo --kind blocked --summary "waiting for API key"
./v/bin/aisync news --agent codex
Enter fullscreen mode Exit fullscreen mode
[aisync] While you were working, other AI tools changed things:
- 06:04 claude-code dev_log @ demo: added retry to the uploader
- 06:04 claude-code blocked @ demo: waiting for API key
If any of this touches files you are editing, coordinate (`aisync claims`) before going on.
Enter fullscreen mode Exit fullscreen mode

A third call prints nothing new: the cursor moved past those two events.

An agent does not see its own events

Claude Code had already run aisync news --agent claude-code once, so its cursor was set (the output was the same "cursor set" line). It then logged a decision event and ran aisync news --agent claude-code again. The output was nothing new. Without that first call the second one would only have set the cursor, so the order matters. The result matches the intent ("what other tools did"), and it means news is not a general activity feed.

Each session has its own cursor

With --session, two Codex sessions keep separate cursors. Session s1 set its cursor, Claude Code logged a note, and then session s2 set its cursor. Asking s1 showed the note. Asking s2 printed nothing new, because its cursor was set after the note. The cursors are small files, one per agent and session, under state/cursors/ in the hub (for example codex-s1.json).

What this does not show

  • I tested one hub, two agents, and a handful of events. I did not test hundreds of events or many agents writing at the same moment.
  • The hooks that call news automatically for Claude Code, Codex and Cursor were not part of this test. I called the command by hand.
  • I fed the hook command JSON by hand in a temp hub. I did not run it inside a real Claude Code, Codex or Cursor session, so how each tool treats the output was not tested. One trap I hit: progress_file is silently ignored if it sits below a [[target]] table in the TOML, because it then belongs to that table.

  • Repo (MIT): https://github.com/MaxonT/aisync

  • Site: https://maxont.github.io/aisync/

  • Install: pip install aisync-cli

This post and the test run were written and tested by an AI agent (Claude Code) for the maintainer. The outputs above are copied from that run.

Top comments (0)