Watch
0
0
Fork
You've already forked hyperhive
0
Commit graph hyperhive/hive-agent-sock
Author SHA1 Message Date
atlas
afe6026d44 subagent: keep runs at +1000 when their cgroup is unavailable
Without a run cgroup (no delegation, setup failed, or one create
failed) a subagent run was started plain at oom_score_adj 0, below the
+500 the unit gave it before. Every run of the daemon now goes through
the entry wrapper; it sets +1000 and enters the cgroup only when there
is one. run_cgroup::entry takes the cgroup as an Option.
2026-10-13 12:25:27 +02:00
atlas
b31db6c7d9 subagent, bash: run each subagent run and bash task in its own cgroup
hive-subagent-daemon and hive-bash-daemon are now delegated their cgroup
(Delegate=memory pids, DelegateSubgroup=daemon). At start each daemon
enables memory+pids below its unit and creates runs/ (subagent) or
tasks/ (bash). Each subagent run gets runs/<name>-<n>/ and each bash
task tasks/<id>/, both with memory.oom.group=1, and the daemon removes
the directory when the run ends. An OOM kill now takes one whole run
instead of one process out of the daemon's shared cgroup, and never the
daemon itself.

Subagent runs get oom_score_adj +1000; bash tasks and both daemons stay
at 0; the harness stays at -500. The unit-level OOMScoreAdjust=500 on
the subagent daemon goes: it was both the daemon's own score and a floor
no run below it could be lowered past.

Bash tasks enter their cgroup from the existing pre_exec hook, writing
to a cgroup.procs fd opened by the parent. hive-claude builds its
Command itself and exposes neither a pre_exec hook nor the child's pid,
so a subagent's claude (and an ACP agent's process) is started as the
daemon's own binary with the run's cgroup in its environment; main()
enters the cgroup, raises oom_score_adj and execs the real program, the
trampoline swarm-review-classifier already uses. Both enter before the
program executes, so nothing it forks starts outside the run's cgroup.

Without a delegated `daemon` leaf (tests, a dev shell) each daemon logs
one warning and runs everything in its own cgroup, as before.

The shared helper is hive-agent-sock's run_cgroup module, the crate
both daemons already use for what in-container plugins share.

Closes #5007
2026-10-13 12:25:27 +02:00
atlas
58ef299c3c hive-c0re: remove the approvals subsystem
Nothing submits an approval since scheduled prompts left hive-c0re, so the
queue only acted on rows written before that.

- hive-c0re: the approvals store and its migrations, approve/deny/
  resolve actions, the ResolveApproval tail node (and meta_update's
  approval_id), POST /api/approve and /api/deny, the approval views on
  /api/state, the approval_resolved dashboard event, the orphan GC,
  destroy/purge failing pending rows, and the approval rows in
  get_loose_ends. DestroyBookkeeping loses the purge flag, which only
  worded the approval-failure note.
- The CancelLooseEnd request goes: its only hive-c0re arm was the
  approval one. cancel_loose_end handles reminder and todo in the harness.
- Wire types: hive_sh4re::approvals (ReminderStats moves to inbox),
  LooseEnd::Approval, CancelLooseEndKind, HelperEvent::ApprovalResolved,
  ToolGroup::Approvals, HostRequest::{Pending,Approve,Deny} and
  HostResponse.approvals. HOST_SOCK_VERSION goes to 2 for the removed
  field.
- hivectl loses the approvals verbs; docs/tools/hivectl-cli.md is
  regenerated.
- The system prompt drops approval_resolved.
- The approvals table stays in broker.sqlite, unread, the same way the
  operator_questions table was left when its store was removed.
- An `approvals` entry still in tool-groups.json is skipped with a warning
  by ToolGroup::parse_list, like any unknown group.
- docs/agent-lifecycle/approvals.md is renamed to config-changes.md so the
  code's doc pointers resolve; its content follows in the docs commit.
2026-10-04 23:55:32 +02:00
atlas
acfa2a7a56 docs(readmes): fix stale facts in remaining crate READMEs + gotchas
- frontend/README.md: list the missing packages/swarm-ui package, fix
  the vanilla-JS claim (all packages depend on preact), and the
  deprecated hyperhive.frontend.extraFiles spelling
- hive-c0re/README.md: hive-c0re no longer provisions per-agent
  forge/matrix accounts (swarm-controller does); it wires gateway
  vhosts and reconciles forge/matrix config
- hive-metric/README.md: OTEL_EXPORTER_OTLP_HEADERS is never set by
  the harness and has no way to be set
- hive-agent-sock/README.md: fix the deprecated
  hyperhive.extraMcpServers spelling
- docs/process/gotchas.md: fix a dangling hive-ag3nt/ path, the real
  directory is hive-agent/

Also fixes pre-existing vale error-level alerts (passive voice,
Microsoft.Auto, Microsoft.Contractions) in the same files so
prose-lint-errors passes clean.
2026-10-02 11:40:07 +02:00
atlas
6ea81aae65 hive-c0re: render the new agent option paths into generated agent flakes
meta.rs writes each agent's flake, and it still named the pre-move
`hyperhive.*` paths — so every agent rebuild would print a rename
deprecation warning about a line no human wrote and no operator could fix.
A warning nobody can act on trains everyone to ignore the ones that matter,
which is the whole value of the alias shims.

Repoints the FORWARDED_VAR_OPTIONS table and every other emitted option
assignment (otel.*, docs.source, claudeCodePath, github.enable, user.name,
claudeMemoryMaxBytes) to `services.hyperhive.agent.*`, with the test
expectations that pin the rendered text. The flake input named `hyperhive`
(`hyperhive.url`, `hyperhive.inputs.nixpkgs.follows`,
`hyperhive.nixosConfigurations.*`), hive-tier `services.hyperhive.*` paths,
and the `@hyperhive.local` git identity share the word and are untouched.

Also repoints the same option paths where they appear in comments, rustdoc
and runtime message strings across the other crates — a refusal message
naming `hyperhive.allowedRecipients` sends an operator to a path that will
stop existing. Prose under docs/ is deliberately not in this commit.

Refs #4473
2026-09-17 20:19:30 +02:00
damocles
349f7096ba subagents: fold hive-extra-mcp into hive-agent-sock per mara's rescoping 2026-09-13 13:57:53 +02:00
atlas
39b95c2ede treefmt: apply prettier
Pure `nix fmt` output from the commit before this one — no hand edits.
203 files: 52 md, 42 tsx, 32 js, 32 css, 21 ts, 13 html, 8 json, 3 mjs.

Reproduce with `nix develop -c nix fmt` on the parent commit; the result
should be byte-identical to this tree.

None of the 13 `.prettierignore` entries appears here — verified by
intersecting the changed-file list against the ignore file, with a
control proving the intersection finds a match when one exists.
2026-09-02 15:25:07 +02:00
damocles
c865114743 hive-agent: remove the in-container questions mirror 2026-08-29 22:27:28 +02:00
damocles
2c7872841a hive-agent: guarantee a wake after a self-requested /compact 2026-08-13 23:17:10 +02:00
damocles
1b72ed56ff todos: reopen an acked keyed row when the caller says so 2026-08-13 12:55:20 +02:00
damocles
80f16094f1 hive-sh4re: split inbox, container, journal, and schedule wire shapes into their own modules
Closes the #3110 split — lib.rs is now just the crate doc comment and
the pub mod list.

journal.rs's new doc comment fixes a pre-existing bug: the old
JournalPriority doc text in lib.rs was actually half Capability's doc
(a leftover from an earlier reorder that moved the code but not the
comment above it).
2026-08-10 23:26:15 +02:00
damocles
02bbff1e34 hive-sh4re: split approval-queue schema into its own topic module 2026-08-10 23:05:18 +02:00
damocles
7ba492b965 move hive-sh4re's harness_dir() into hive-agent-sock, drop the dev-fallback derivation 2026-08-09 19:54:13 +02:00
damocles
642377c5e0 docs: stop implying matrix/bash/forge is a closed todo-producer list 2026-08-04 00:06:19 +02:00
damocles
bc69ee3b8f revise bulk-clear to explicit ids per mara's feedback, fix clippy line count 2026-08-02 16:41:33 +02:00
damocles
eb570c003d cap get_loose_ends todo output + add ack_todos_until bulk-clear (#2944) 2026-08-02 16:41:33 +02:00
damocles
e5ef5a72be feat(#2635): wire harness-local questions mirror (inc2 pt2) 2026-07-26 02:07:33 +02:00
atlas
bbfc578c95 docs(#2627): add READMEs for the remaining infra crates
Second increment of the per-crate README effort, covering the rest of the
infra/wire/priv column: hive-priv, hive-metric, hive-types, hive-sh4re,
hive-core-agent-sock, hive-agent-sock. Same shape as the first batch —
purpose + when-to-use, and point at the crate-root //! docs plus the
relevant docs/ pages rather than duplicating them. Wires
readme = "README.md" into each Cargo.toml [package].

Disjoint from the batch-1 crates, so the two increments compose cleanly.
2026-07-23 13:16:29 +02:00
damocles
e9df5def02 add agent-facing compact tool gated on context usage 2026-07-23 12:34:20 +02:00
damocles
dcb2b79715 strip issue-tag comments from #2635 inc1 commits per hive-rules 2026-07-23 00:12:30 +02:00
damocles
ecd9030305 wire reminder ops into the in-agent socket dispatch (#2635 inc 1) 2026-07-23 00:12:30 +02:00
damocles
18d21237bf add reminder ops to the in-agent socket protocol (#2635 inc 1) 2026-07-23 00:12:30 +02:00
damocles
86d16efa07 feat(#2569): drive a turn on a local todo signal (serve-loop select) 2026-07-20 23:29:26 +02:00
damocles
d85e1895dc feat(#2569): add the in-agent hive-agent-sock crate (todo wire types) 2026-07-20 23:29:26 +02:00
damocles
795dd882bb refactor(#2569): rename hive-agent-sock to hive-core-agent-sock 2026-07-20 21:58:28 +02:00
damocles
144912f8e0 address review: drop backwards-compat request/response aliases, use canonical names 2026-07-19 15:53:09 +02:00
damocles
d0beec8a40 refactor(#2581): carve per-agent mcp.sock protocol into hive-agent-sock crate 2026-07-19 15:53:09 +02:00