refactor: rename pod crate to worker
This commit is contained in:
@@ -7,7 +7,7 @@ p.feature = {
|
||||
task = { enabled = true },
|
||||
memory = { enabled = true },
|
||||
web = { enabled = true },
|
||||
pods = { enabled = false },
|
||||
workers = { enabled = false },
|
||||
ticket = { enabled = false, access = "lifecycle" },
|
||||
ticket_orchestration = { enabled = false },
|
||||
}
|
||||
|
||||
@@ -6,7 +6,7 @@ p.feature = {
|
||||
task = { enabled = true },
|
||||
memory = { enabled = true },
|
||||
web = { enabled = true },
|
||||
pods = { enabled = true },
|
||||
workers = { enabled = true },
|
||||
ticket = { enabled = true, access = "lifecycle" },
|
||||
ticket_orchestration = { enabled = false },
|
||||
}
|
||||
|
||||
@@ -7,7 +7,7 @@ return yoi.profile {
|
||||
record_event_trace = true,
|
||||
},
|
||||
|
||||
worker = {
|
||||
engine = {
|
||||
reasoning = "high",
|
||||
},
|
||||
|
||||
@@ -23,7 +23,7 @@ return yoi.profile {
|
||||
task = { enabled = true },
|
||||
memory = { enabled = true },
|
||||
web = { enabled = true },
|
||||
pods = { enabled = true },
|
||||
workers = { enabled = true },
|
||||
ticket = { enabled = false, access = "lifecycle" },
|
||||
ticket_orchestration = { enabled = false },
|
||||
},
|
||||
|
||||
@@ -7,7 +7,7 @@ p.feature = {
|
||||
task = { enabled = false },
|
||||
memory = { enabled = true },
|
||||
web = { enabled = true },
|
||||
pods = { enabled = false },
|
||||
workers = { enabled = false },
|
||||
ticket = { enabled = true, access = "lifecycle" },
|
||||
ticket_orchestration = { enabled = false },
|
||||
}
|
||||
|
||||
@@ -7,7 +7,7 @@ p.feature = {
|
||||
task = { enabled = false },
|
||||
memory = { enabled = true },
|
||||
web = { enabled = true },
|
||||
pods = { enabled = true },
|
||||
workers = { enabled = true },
|
||||
ticket = { enabled = true, access = "lifecycle" },
|
||||
ticket_orchestration = { enabled = true },
|
||||
}
|
||||
|
||||
@@ -7,7 +7,7 @@ p.feature = {
|
||||
task = { enabled = false },
|
||||
memory = { enabled = true },
|
||||
web = { enabled = true },
|
||||
pods = { enabled = false },
|
||||
workers = { enabled = false },
|
||||
ticket = { enabled = false, access = "lifecycle" },
|
||||
ticket_orchestration = { enabled = false },
|
||||
}
|
||||
|
||||
+4
-4
@@ -1,12 +1,12 @@
|
||||
---
|
||||
## Pod orchestration
|
||||
## Worker orchestration
|
||||
|
||||
When Pod-management tools are available, spawned Pod notifications are background signals for the parent to handle at a natural stopping point. Do not ignore routine follow-up, but do not interrupt the current user request unnecessarily.
|
||||
When Worker-management tools are available, spawned Worker notifications are background signals for the parent to handle at a natural stopping point. Do not ignore routine follow-up, but do not interrupt the current user request unnecessarily.
|
||||
|
||||
The parent does not need to keep a turn open or call tools solely to wait for a notification. Do not use `sleep` or polling loops just to wait for Pod output; if there is no useful immediate work, return control and handle the child when notified or when the user next asks.
|
||||
The parent does not need to keep a turn open or call tools solely to wait for a notification. Do not use `sleep` or polling loops just to wait for Worker output; if there is no useful immediate work, return control and handle the child when notified or when the user next asks.
|
||||
|
||||
Before treating delegated work as complete, read the child output and inspect concrete evidence such as worktree state, diff, and test results. Notifications are hints, not proof of completion.
|
||||
|
||||
Peer Pods made visible by reciprocal metadata registration are not spawned children. Use peer messaging only as explicit communication; it does not grant scope, produce a child output cursor, imply parent ownership, or create child completion notifications. Peer sends require a live peer and do not auto-restore stopped peers.
|
||||
|
||||
This guidance is not scheduler or auto-maintain authorization. Do not start workflows, merge or clean up work, close tickets, or bypass user/workflow authorization solely because Pod tools or notifications exist.
|
||||
This guidance is not scheduler or auto-maintain authorization. Do not start workflows, merge or clean up work, close tickets, or bypass user/workflow authorization solely because Worker tools or notifications exist.
|
||||
@@ -1,11 +1,11 @@
|
||||
# Pod internal prompts (builtin pack).
|
||||
# Worker internal prompts (builtin pack).
|
||||
#
|
||||
# Values are minijinja template strings. Use `{% include "$prefix/..." %}`
|
||||
# to pull in long text from the $yoi / $user / $workspace prompt
|
||||
# libraries.
|
||||
#
|
||||
# Every key here MUST correspond to a `PodPrompt` variant; missing or
|
||||
# extra keys cause a build-time error (see `crates/pod/build.rs`).
|
||||
# Every key here MUST correspond to a `WorkerPrompt` variant; missing or
|
||||
# extra keys cause a build-time error (see `crates/worker/build.rs`).
|
||||
|
||||
[prompt]
|
||||
compact_system = "{% include \"$yoi/internal/compact_system\" %}"
|
||||
@@ -66,19 +66,19 @@ The following workflows are advertised resident. When a user request matches one
|
||||
{{ entries }}\
|
||||
"""
|
||||
|
||||
pod_orchestration_guidance_section = "{% include \"$yoi/common/pod-orchestration\" %}"
|
||||
worker_orchestration_guidance_section = "{% include \"$yoi/common/worker-orchestration\" %}"
|
||||
|
||||
ticket_event_companion_notice = "{% include \"$yoi/pod/ticket_event_companion_notice\" %}"
|
||||
ticket_event_companion_notice = "{% include \"$yoi/worker/ticket_event_companion_notice\" %}"
|
||||
|
||||
spawn_pod_tool_description = """\
|
||||
Spawn a new Pod process to work on a delegated task. The spawner's write scope is reduced by the scope passed here; the spawned Pod receives its own socket and starts running `task` immediately. The spawned Pod outlives the spawner's current turn and can be contacted again through its socket path.
|
||||
spawn_worker_tool_description = """\
|
||||
Spawn a new Worker process to work on a delegated task. The spawner's write scope is reduced by the scope passed here; the spawned Worker receives its own socket and starts running `task` immediately. The spawned Worker outlives the spawner's current turn and can be contacted again through its socket path.
|
||||
|
||||
Optional `cwd`: when provided, it is the child process/tool default working directory only. It must be an absolute existing directory covered by the child's delegated readable scope, and it does not change workspace/Profile/memory/Ticket roots or grant authority.
|
||||
|
||||
Profile selection: `profile` may be omitted or set to `default` to use the effective child default profile, set to `inherit` to derive reusable child configuration from this Pod, or set to one of the registry selectors below. Raw/path profile selectors are not accepted by SpawnPod. `scope` is always the only delegated filesystem capability; profile scope is replaced by the explicit SpawnPod scope.
|
||||
Profile selection: `profile` may be omitted or set to `default` to use the effective child default profile, set to `inherit` to derive reusable child configuration from this Worker, or set to one of the registry selectors below. Raw/path profile selectors are not accepted by SpawnWorker. `scope` is always the only delegated filesystem capability; profile scope is replaced by the explicit SpawnWorker scope.
|
||||
|
||||
Default profile: {{ default_profile }}
|
||||
Special selector: inherit — derive reusable model/worker/tool policy from the spawner while replacing pod.name and scope.
|
||||
Special selector: inherit — derive reusable model/worker/tool policy from the spawner while replacing worker.name and scope.
|
||||
Available registry profiles:
|
||||
{{ available_profiles }}{% if profile_diagnostic %}
|
||||
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
<system-reminder>
|
||||
Workspace Dashboard observed that this Orchestrator Pod is idle while queued Ticket work is present.
|
||||
Workspace Dashboard observed that this Orchestrator Worker is idle while queued Ticket work is present.
|
||||
|
||||
This is bounded attention only, not scheduler authority. Do not drain the queue automatically. Before implementation side effects, verify the Ticket state and record the normal `queued -> inprogress` acceptance through Ticket tools.
|
||||
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
---
|
||||
description: worktree と sibling の coder / reviewer Pod を使い、下位 orchestrator が concrete Ticket 群の実装・外部レビュー・修正・完了準備を管理する orchestration フロー
|
||||
description: worktree と sibling の coder / reviewer Worker を使い、下位 orchestrator が concrete Ticket 群の実装・外部レビュー・修正・完了準備を管理する orchestration フロー
|
||||
model_invokation: true
|
||||
user_invocable: true
|
||||
requires: [workflow-resource-boundary]
|
||||
@@ -11,4 +11,4 @@ requires: [workflow-resource-boundary]
|
||||
2. Coder は local tactics を選んで実装し、変更ファイル・summary・validation・commit/status・unresolved notes・review readiness を報告する。design/permission/history/prompt-context/dependency/Ticket-boundary の未決事項は Orchestrator に escalate する。
|
||||
3. Reviewer は Ticket/intent packet、binding decisions/invariants、acceptance criteria、diff/commits、validation evidence を読んでから判断する。実装を recorded requirements と project design boundary に照らして review し、blocker / non-blocking follow-up / parent-decision item を分類する。
|
||||
4. Reviewer report は approve/request-changes evidence、file/line evidence where useful、validation performed、risks、unresolved concerns を含める。Reviewer は merge、close、push、cleanup、root/original workspace operation、Coder への直接指示を行わない。
|
||||
5. Orchestrator は blocker fix loop、review approval、merge-ready dossier、integration/cleanup/close authority を管理する。下位 Pod の completion notification は transient hint とし、durable output/session/worktree state で確認する。
|
||||
5. Orchestrator は blocker fix loop、review approval、merge-ready dossier、integration/cleanup/close authority を管理する。下位 Worker の completion notification は transient hint とし、durable output/session/worktree state で確認する。
|
||||
|
||||
@@ -14,7 +14,7 @@ Intake の目的は、曖昧な依頼をいきなり実装委譲せず、Orchest
|
||||
|
||||
Intake は以下をしない。
|
||||
|
||||
- coder / reviewer / read-only investigation helper Pod を起動しない。
|
||||
- coder / reviewer / read-only investigation helper Worker を起動しない。
|
||||
- implementation worktree を作らない。
|
||||
- implementation / review routing、merge、close、branch cleanup をしない。
|
||||
- unattended scheduler として自動実行しない。
|
||||
|
||||
@@ -13,4 +13,4 @@ requires: [workflow-resource-boundary]
|
||||
4. child implementation worktree は recorded implementation worktree root 配下に作成し、implementation branch は Orchestrator workspace current HEAD / orchestration branch HEAD から切る。Coder/Reviewer は sibling Pods として route し、Ticket intent・binding decisions・validation/report expectations を concrete handoff として渡す。
|
||||
5. Reviewer が request-changes した場合は、scope 内で Coder に修正を戻すか、Orchestrator が blocked/escalation を記録する。approve evidence が揃うまで merge-ready とみなさない。
|
||||
6. merge completion authority が無い場合は、Ticket/branch/commit identity、intent/invariant check、implementation summary、coder/reviewer evidence、blocker resolution、validation、risks、dirty state、decision needs を含む merge-ready dossier で停止する。
|
||||
7. 明示 authority と clean reviewer evidence がある場合のみ、merge target workspace への統合、validation、Ticket close/completion processing、関連 Pod 停止、scope reclamation、worktree/branch cleanup、evidence reporting を進める。
|
||||
7. 明示 authority と clean reviewer evidence がある場合のみ、merge target workspace への統合、validation、Ticket close/completion processing、関連 Worker 停止、scope reclamation、worktree/branch cleanup、evidence reporting を進める。
|
||||
|
||||
Reference in New Issue
Block a user