Commit graph hyperhive/hive-subagent-mcp/src/session.rs
Author SHA1 Message Date
atlas
ae84174e3a subagent daemon: explain the status when returning it, not up front
The `status` tool description enumerated all five states it can report, so
every caller paid for four answers it didn't get and read the explanation
in the wrong place. The description now states only what the tool is for
and that it costs nothing to call; each answer it returns carries its own
meaning and the caller's next move instead — running, starting and idle
were terse, the not-found error terser still, and they had been leaning on
the enumeration to be legible. The killed answer and the end-of-turn todo
are unchanged. `continue`'s description keeps its killed-resume sentence:
that describes what the tool does, not a state it might hand back.

Refs #4326
2026-09-13 15:23:25 +02:00
atlas
0bdee751b9 subagent daemon: address review on the killed-session docs
argus's review on #4333 flagged one real vale error (Microsoft.Contractions):
"did not" in the new "A killed turn" section reads as "didn't" instead. The
rest of the diff's new prose (docs/tools/subagent.md and the tool
descriptions in hive-subagent-mcp/src/mcp.rs) has no other instance of the
same expansion, confirmed both by a local `vale --minAlertLevel=error` run
(clean) and by grepping the added lines. Also tidies session.rs's module doc:
the two `//!` runs split to dodge the 30-line comment-block lint had a bare
trailing `//!` right before the blank separator; dropped so the first run
ends on content.

Refs #4326
2026-09-13 15:15:38 +02:00
atlas
f817e27d4c subagent daemon: report a killed session as killed, not idle
A subagent whose claude process died on a signal — the kernel's OOM
killer, a stopped unit, an `interrupt` — was indistinguishable from one
that finished its turn: its entry left the `running` map, `status` fell
through to "a session exists on disk" and answered `idle`, and the
end-of-turn todo said the subagent had "finished". The usual next move
on that reading is `continue`, which resumes work that was cut mid-turn
with nothing having recorded that it was cut.

The driver already preserves how the child ended — `RunningClaude::wait`
returns `Error::Exit` carrying the `ExitStatus`, whose `signal()` is the
whole answer — so this reads it rather than having to recover it:
`classify_end` turns the outcome into `Complete` / `Killed { signal }` /
`Failed`, and `State::finish_turn` remembers a kill against the name
(cleared by the next confirmed spawn under it).

What an agent sees as a result:

- `status` reports the session killed, naming the signal, instead of idle;
- the todo the daemon pushes without being asked says the subagent was
  KILLED mid-turn rather than that it finished;
- `continue` still resumes such a session, but its reply says the
  previous turn was killed, so no caller carries on from cut-off work
  believing it was complete.

Refs #4326
2026-09-13 14:58:45 +02:00
damocles
16eec3c314 subagents: add availableToSubagents opt-in toggle for extraMcpServers 2026-09-13 13:57:53 +02:00
damocles
101a786dff hive-subagent-mcp: fix resolve_dir commit ordering + add non-committing peek_dir 2026-09-12 01:48:55 +02:00
damocles
25499ce8df subagent: note the name-only concurrency guard now that dir adds a second axis 2026-09-12 01:48:55 +02:00
damocles
aa27334d76 subagent tool: remember dir per session name so continue/status don't need to repeat it 2026-09-12 01:48:55 +02:00
damocles
cfddb1bd40 subagent tool: add option to run in a different directory 2026-09-12 01:48:55 +02:00
damocles
d26b754701 subagent: fix a real test race on the process-wide OTEL_RESOURCE_ATTRIBUTES env var 2026-09-09 23:47:21 +02:00
damocles
1ad7145cc5 swap hive-claude back to the published 0.1.1 release now that it's out 2026-09-09 23:45:12 +02:00
damocles
561bd09618 subagent: close the start/continue TOCTOU race with an atomic reservation 2026-09-09 23:45:12 +02:00
damocles
c2fb3c6e3e subagent: add status tool, cut docs down to operator-facing + no cli flags 2026-09-09 23:45:12 +02:00
damocles
7699db6500 hive-subagent-mcp: new crate for the subagent daemon, independent of hive-bash-mcp 2026-09-09 23:45:12 +02:00