vinzenz/hyperhive
vinzenz/hyperhive
Watch
0
0
Fork
You've already forked hyperhive
 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
hyperhive/hive-agent
History
Repository files (latest commit first)
Filename Latest commit message Latest commit date
atlas 7ba90d5a3f feat(agent): raise a todo when the agent's disk gets tight 
Closes 2718.

The operator has been going through agent dirs by hand with ncdu,
deleting 20+GB target dirs. Agents had no way to know they were the
ones sitting on the space.

New `disk_watch` module in the harness: every 15 minutes it statvfs's
the filesystem backing the agent's state dir and, past 80%, raises a
keyed `disk` todo telling the agent to free space — with the operator's
rules inline: only delete things that are actually big, build output
first, and never delete something still needed, ask for more space
instead.

Over threshold it also walks the agent's own tree (`/agents/<label>`
plus `$HOME`) and names the directories worth looking at, so the todo
says where the bytes actually went rather than just that the disk is
full. The walk is bounded on every axis — entry budget, recursion cap,
report depth — pinned to the state dir's device so it can't wander into
`/nix` or the shared bind mounts, and it does not traverse symlinks. It
reports the deepest oversized directory on each branch, so the agent
gets pointed at `<workspace>/target` rather than at `/agents/<label>`.

Anti-nag is the whole design constraint. The todo is keyed, and the
summary is deliberately stable: the percentage is bucketed to 5 points
and no raw byte counts appear anywhere in it. An unchanged situation
re-upserts as `changed == false` and never fires the wake, so a disk
that has been steady at 89% for a week sits quietly in the loose-ends
list; crossing into a new bucket speaks up once. Dropping back under
the threshold clears the row.

Harness-local by construction, per the operator's call that this gets
no core wiring: hive-c0re cannot push a todo at all (the store and its
wake live inside the container), and running in-process means this
skips even the in-agent socket and calls `Todos::upsert` directly.

Worth recording, since it shaped the scope: btrfs does NOT fold qgroup
limits into statfs. Measured with quota counting enabled and a 20G
limit set on a real subvolume, statvfs returns byte-identical whole-FS
numbers for that subvolume, an ordinary agent dir, and the root. So
this watches host-FS pressure, which is valid before and after the
planned subvolume migration; per-agent quota awareness would need the
limit handed to the agent explicitly.
2026-07-26 17:59:13 +02:00
..
prompts prompts: fix rebase conflict + qualify inbox tool names consistently 2026-07-26 17:33:41 +02:00
src feat(agent): raise a todo when the agent's disk gets tight 2026-07-26 17:59:13 +02:00
Cargo.toml feat(agent): raise a todo when the agent's disk gets tight 2026-07-26 17:59:13 +02:00
README.md remove hive-agent-wake — no shipped consumer 2026-07-25 20:05:32 +02:00

README.md

hive-agent

The in-container harness serve-loop binary — one instance per agent. Long-polls the broker inbox and drives one claude --print turn per inbox message, over the hive-claude driver. There is one role here (agent); the Surface trait + AgentSurface zero-sized type tag keep the turn loop generic and testable for future roles without a parallel copy of the loop.

When to use it

You don't call into this crate from elsewhere — it's the top-level binary systemd starts per agent container. Look here when you need to understand or change: what happens between "a message lands in the inbox" and "claude produces a reply", how login/auth is bootstrapped, how the per-agent web UI is served, or how turn/event stats get recorded. Architecture detail lives in docs/turn-loop.md; this README is just the map of the module tree.

Shape

  • turn.rs — the turn-loop policy layer: renders the system prompt + MCP config, invokes hive-claude, classifies the outcome, and feeds the event/turn-stats sinks.
  • client.rs — broker client (inbox poll, ack, send) speaking the hive-sh4re wire protocol.
  • login.rs / login_session.rs — first-run and session-resume auth flow for the claude CLI.
  • mcp_config.rs — renders the per-turn --mcp-config / --allowedTools blob from tool groups + capabilities.
  • todos.rs / reminders.rs / todo_server.rs — the harness-local loose-ends v2 stores (sqlite-backed) and the in-agent socket server extra MCP daemons + hive-agent-mcp dial into for todo/reminder ops.
  • vacuum.rs — periodic sqlite vacuum sweep for the harness-local stores.
  • events.rs / turn_stats.rs / stats.rs — append-only event sink and per-turn telemetry recording (context usage, cost, tool favorites) under harness/.
  • forge_notify.rs — subscribes to forge notifications and wakes the harness on new activity.
  • prompt.rs — system-prompt renderer (persona + tool docs + environment facts).
  • web_ui/ — the per-agent dashboard (terminal pane, status, schedules) served over the built-in hive-agent web port.
  • paths.rs — canonical path resolution for state/harness dirs and the harness-local sqlite files.

Sibling: hive-agent-mcp (the MCP server this loop points claude at every turn). Both are described together in docs/turn-loop.md::Harness binary shape.