docs: clear write-good.Passive hits in docs/agent-lifecycle/agent-hierarchy.md

3 genuine passive-voice rewrites (actor named — the operator,
hive-c0re, the harness); 5 false positives left untouched — a
design-intent idiom ('are meant to run'), a predicate-adjective state
description ('is finished'), a predicate-adjective necessity
description ('is required', the original established example for this
category), a project-planning-state description ('is planned'), and a
quoted literal error string ('must be owned by uid 0...').
This commit is contained in:
iris 2026-09-20 05:22:46 +02:00 committed by mara
commit ad16f3dba5

View file

@ -20,9 +20,9 @@ Topology lives in the hive-c0re-owned **meta repo**, alongside
```
`null` = root-level agent. New agents **default to root** — there is no
structural manager that everything hangs under. Hierarchy is built
explicitly: an agent gets a parent edge written before its first spawn,
or the operator reparents it afterwards (so `bob` above sits under
structural manager that everything hangs under. The operator builds
hierarchy explicitly: an agent gets a parent edge written before its
first spawn, or the operator reparents it afterwards (so `bob` above sits under
`alice`). Any agent is reparentable, the bootstrap container (`ruth`)
included — it's just another root. The manager is reparentable like any other agent; there's no
"structurally root" carve-out. Its privileges live on its MCP socket,
@ -128,8 +128,8 @@ other agents don't:
`/applied` (diff against what's deployed) and `/meta` (system-wide
deploy log). Planned: each agent gets RW to `/agents/<descendant>/`
for just its own subtree — the manager's full-forest RW becomes the
"root's subtree is everything" case of that same rule. RO `/meta`
access will be gated on a "meta read" capability; only
"root's subtree is everything" case of that same rule. hive-c0re will
gate RO `/meta` access on a "meta read" capability; only
`request_update_meta_inputs` writes `flake.lock`, gated by its own
capability.
- **Prompt/tools** — the system prompt uses `<!-- role:agent -->` /
@ -151,8 +151,8 @@ Manager}` switch picks the MCP tool allow-list claude sees. Both are
ancestor/descendant check instead of a manager-name check — see the
module docs for `loose_ends.rs`, `stores/broker.rs`, `actions.rs`,
and `workers/crash_watch.rs` for the current owner-check logic in
each. (Reminder cancellation is handled fully in-agent — see the note
on `CancelLooseEndKind::Reminder` in
each. (The harness handles reminder cancellation fully in-agent — see
the note on `CancelLooseEndKind::Reminder` in
`hive-c0re/src/socket_server/mod.rs`.)
None of the above is a stable interface — treat the module doc