docs(agent-hierarchy): mark section D done — legacy /state mount dropped (#604)

This commit is contained in:
lexis 2026-05-29 22:08:14 +02:00 committed by Mara
commit d074b25fac

View file

@ -145,14 +145,16 @@ Tree-shape version:
- `request_update_meta_inputs` is the only path that actually writes - `request_update_meta_inputs` is the only path that actually writes
`flake.lock`, gated by the cap; everyone else stays RO. `flake.lock`, gated by the cap; everyone else stays RO.
### D — drop legacy `/state` for manager ### D — drop legacy `/state` for manager ✓ done (#604)
`lifecycle.rs::notes_mount` currently ternary's `/state` for the `lifecycle.rs` no longer binds `/state` for the manager.
manager and `/agents/<name>/state` for everyone else (because the `HYPERHIVE_STATE_DIR` is now injected uniformly via
manager pre-dates the per-agent state-dir layout). Milestone bullet: `systemd.globalEnvironment` in `meta.rs` for every container
unify on `/agents/<name>/state` for everyone. One-time `mv` of (manager included), so all token/state paths resolve through
`/var/lib/hyperhive/manager/state``/var/lib/hyperhive/agents/manager/state` `$HYPERHIVE_STATE_DIR`. The harness-base shell scripts
in `migrate.rs` (idempotent, marker-guarded). (tea-login, forge-avatar-sync, matrix-avatar-sync) simplified
from glob+for loops to a direct `$HYPERHIVE_STATE_DIR/<token>`
read.
### E — prompt + tools ### E — prompt + tools