Design: lean LXC dispatcher with remote-only terminals (P1-6 generic remote launch)

A paused terminal gets no new runs and is not probed; in-flight runs still finish.

agentd/cli.py

@@ -116,6 +116,9 @@ def cmd_smoke(cfg, a):
     body = {"name": "agentd smoke %s %s" % (a.terminal, time.strftime("%Y-%m-%d %H:%M")),
             "description_html": SMOKE_BRIEF, "state": pl.states(p["id"])[p["states"]["todo"]],
             "assignees": [pl.members()[user]]}
+    if a.labels:
+        labels = pl.labels(p["id"])
+        body["labels"] = [labels[x] for x in a.labels.split(",")]
     issue = pl.create_issue(p["id"], body)
     key = "%s-%s" % (p.get("key_prefix", a.project), issue["sequence_id"])
     print("created %s; ticking every %ss (timeout %s min)" % (key, a.interval, a.timeout))
@@ -195,6 +198,7 @@ def parser():
     s.add_argument("terminal")
     s.add_argument("--interval", type=int, default=30)
     s.add_argument("--timeout", type=int, default=30, help="minutes")
+    s.add_argument("--labels", default="", help="comma-separated Plane labels, e.g. model:sol")
     s.add_argument("--as", dest="as_")
     s.set_defaults(fn=cmd_smoke)
     sub.add_parser("check-config", help="validate the config and list token env vars").set_defaults(fn=cmd_check_config)

config/agentd.yaml

@@ -0,0 +1,82 @@
+# plane-agentd live configuration on oregon (phase 1). No secrets: token_key values are Vault key names; the tokens
+# are injected by `agent-access.py exec --secret-env KEY=KEY` (see cron/agentd-tick.sh and docs/runbook.md).
+plane: {base_url: https://plane.kappiee.com, workspace: kappiee}
+vault: {client: ~/.local/share/server-access-client/agent-access.py, config: https://nas.tail9531b.ts.net:8443/identity/agent.json}
+
+agentd:
+  state_dir: ~/.gwent-board/agentd     # JSON state, agentd.log, heartbeat.json, local run dirs (runs/)
+  helpers_dir: /home/ubuntu/Documents/wjj/project/gwent-design/tools   # win10-ps.sh / win10-put.sh
+  interval: 120
+  max_rounds_per_thread: 8
+  comment_max_chars: 6000
+  cli_terminal: oregon
+
+terminals:
+  win10:
+    transport: warpgate-ssh
+    target: win10
+    os: windows
+    launch: schtasks
+    repo: 'C:\Users\Administrator\Documents\GwentAssistantProject\gwent'
+    run_root: 'C:\Users\Administrator\Documents\GwentAssistantProject\gwent\runtime\agentd'
+    bins: {codex: 'C:\Users\Administrator\AppData\Local\Programs\OpenAI\Codex\bin\codex.exe'}
+    plane_user: window10
+    token_key: PLANE_WINDOW10_API_TOKEN
+    concurrency: 1
+    busy_processes: [codex]             # do not start while a codex.exe agentd did not start is running
+    capabilities: [android, game]
+    executors: [codex]
+  oregon:
+    transport: local
+    os: linux
+    launch: subprocess
+    repo: /home/ubuntu/Documents/wjj/project/gwent
+    plane_user: oregon
+    token_key: PLANE_OREGON_API_TOKEN
+    concurrency: 1                      # one shared worktree: no parallel edits
+    bins: {claude: /home/ubuntu/.local/bin/claude}   # cron has a minimal PATH; codex is not installed on oregon
+    executors: [claude-code]
+
+executors:
+  codex:
+    adapter: codex
+    run: "codex exec --json -C '{repo}' -s workspace-write -m {model} -c 'model_reasoning_effort=\"{effort}\"' {tier} -o '{out}' -"
+    # `exec resume` takes no -C/-s: the runner already cd's into the repo, the sandbox goes in as a -c override
+    resume: "codex exec resume {thread} --json -m {model} -c 'model_reasoning_effort=\"{effort}\"' -c 'sandbox_mode=\"workspace-write\"' {tier} -o '{out}' -"
+    tier_arg: "-c 'service_tier=\"{tier}\"'"
+    thread_from: events.jsonl:thread.started.thread_id
+    last_message: '{out}'
+    models: [gpt-6.1-sol, gpt-6-astra]
+    default_effort: medium
+  claude-code:
+    adapter: claude-code
+    run: "claude -p --model {model} --output-format json --permission-mode {perm}"
+    resume: "claude -p --resume {thread} --model {model} --output-format json --permission-mode {perm}"
+    permission_mode: acceptEdits
+    thread_from: json:session_id
+    last_message: json:result
+    models: [sonnet, opus, haiku]
+
+projects:
+  GWENT:
+    id: 79cb5a60-aa4f-4138-9864-1dd70426e665
+    key_prefix: GW
+    states: {backlog: Backlog, todo: Todo, running: In Progress, waiting: Waiting, review: Review, done: Done}
+    reviewer: oregon
+    answerer: oregon
+    protocol_file: /home/ubuntu/Documents/wjj/project/gwent/design/board-protocol.md
+    routing:
+      by_assignee: {window10: win10, oregon: oregon}
+      by_label:
+        'model:astra': {model: gpt-6-astra, effort: high}
+        'model:sol': {model: gpt-6.1-sol, tier: fast}
+        'effort:high': {effort: high}
+        'needs-game': {serialize_group: game}
+      default_executor: {win10: codex, oregon: claude-code}
+    retry:
+      transient_max: 3
+      backoff_minutes: [2, 5, 10]
+    timeout_hours: 4
+    max_rounds_per_thread: 8
+    max_progress_rounds: 12
+    # notify (ntfy) is phase 3: left out on purpose

config/schema.json

@@ -53,7 +53,7 @@
         "repo": {"type": "string"},
         "plane_user": {"type": "string"},
         "token_key": {"type": "string"},
-        "concurrency": {"type": "integer", "minimum": 1},
+        "concurrency": {"type": "integer", "minimum": 0},
         "capabilities": {"$ref": "#/definitions/strings"},
         "executors": {"type": "array", "items": {"type": "string"}, "minItems": 1},
         "bins": {"type": "object", "additionalProperties": {"type": "string"}},

cron/agentd-tick.sh

@@ -0,0 +1,9 @@
+#!/usr/bin/env bash
+# cron entry (*/2): one agentd tick under server-access exec (Plane tokens in the process env only), with a lock.
+# The outer flock skips a tick while the previous one still runs; agentd also holds its own state lock.
+exec flock -n /home/ubuntu/.gwent-board/agentd.lock \
+  timeout 900 python3 -I /home/ubuntu/.local/share/server-access-client/agent-access.py \
+  --config https://nas.tail9531b.ts.net:8443/identity/agent.json exec \
+  --secret-env PLANE_OREGON_API_TOKEN=PLANE_OREGON_API_TOKEN \
+  --secret-env PLANE_WINDOW10_API_TOKEN=PLANE_WINDOW10_API_TOKEN --timeout 850 -- \
+  python3 /home/ubuntu/Documents/wjj/project/plane-agentd/cli/agentd tick --once >> /home/ubuntu/.gwent-board/agentd-cron.log 2>&1

docs/design.md

@@ -94,4 +94,5 @@ docs/              design.md executor-contract.md runbook.md
 ## 9. 部署位置
 - Plane VM（plane-01）只跑 Plane：它是设备化节点（固定版本、每日备份停写路径），不放 agentd。
 - 阶段 1–2：agentd 跑在 oregon（已有 Vault 客户端、Warpgate 中继脚本、Claude Code、现有 worker 状态），cron 拉起；配置页面同进程，只监听 tailnet 地址。
-- 阶段 3 起：迁到 PVE 上的专用小节点 `agentd-01`（1–2 vCPU / 2 GB，tailnet + 内网 10.77.10.0/24，与 Plane/Temporal 同网段，webhook 走内网），装 Vault 客户端、中继脚本、Claude Code 与 Codex CLI，使其同时也是一个执行终端（跑核对/验收类任务）。对外访问经 Warpgate HTTP target，复用既有登录。
+- 阶段 3 起：迁到 PVE 上的精简 LXC `agentd-01`（1 vCPU / 1–2 GB，tailnet + 内网 10.77.10.0/24，与 Plane/Temporal 同网段，webhook 走内网）。**LXC 上不装任何 agent CLI**，只有 Python 3.10+、PyYAML、Agent Vault 客户端、ssh 客户端；所有终端（win10、oregon、mac…）一律远程启动（`transport: warpgate-ssh`），LXC 只负责调度、回帖、通知。对外访问经 Warpgate HTTP target，复用既有登录。
+- 远程启动的通用实现（P1-6）：Windows 用计划任务（已实现）；Linux/macOS 经 Warpgate SSH 上传 `run.sh`，`setsid nohup` 脱离会话，轮询 `exit` 文件取回 `last.md`/线程 ID；中继目标名由终端配置给出，不写死。

docs/runbook.md

@@ -40,6 +40,7 @@ cli/agentd comment GWENT 12 reply.md # 或 - 读 stdin
 cli/agentd state GWENT 12 Todo       # Plane 状态名或配置键（todo/review…）
 cli/agentd assign GWENT 12 window10
 cli/agentd smoke GWENT oregon        # 建一个要求回复 "[review] ok" 的工作项，tick 到结束，PASS 则移到 Done
+cli/agentd smoke GWENT win10 --labels model:sol   # --labels 给冒烟项加标签（路由到模型/档位）
 ```
 CLI 以 `--as <terminal>` / `agentd.cli_terminal` / 第一个有 token 的终端的身份操作。
 

tests/test_config.py

@@ -45,6 +45,12 @@ class ConfigTest(unittest.TestCase):
         d["terminals"]["oregon"]["concurrency"] = "two"
         self.check_both(d, "concurrency")
 
+    def test_concurrency_zero_is_valid(self):  # 0 = terminal paused (nothing new is dispatched)
+        d = config_data()
+        d["terminals"]["win10"]["concurrency"] = 0
+        self.assertEqual(config.validate(d), [])
+        self.assertEqual(config.mini_validate(d, SCHEMA), [])
+
     def test_semantic_routing_to_unknown_terminal(self):
         d = config_data()
         d["projects"]["GWENT"]["routing"]["by_assignee"]["mac"] = "mac"

tests/test_core.py

@@ -160,6 +160,14 @@ class CoreTest(unittest.TestCase):
         self.assertEqual(started, ["GW-1-1", "GW-3-1"])  # 2 slots; GW-2 waits for the game group
         self.assertEqual([self.p.state_of(x) for x in (a, b, c, d)], ["In Progress", "Todo", "In Progress", "Todo"])
 
+    def test_concurrency_zero_pauses_terminal(self):
+        self.h.cfg.terminals["win10"]["concurrency"] = 0
+        w = self.p.add_issue(1)
+        o = self.p.add_issue(2, assignee="oregon")
+        self.h.tick()
+        self.assertNotIn(("win10", "codex"), self.h.execs)  # not even probed (busy_processes)
+        self.assertEqual((self.p.state_of(w), self.p.state_of(o)), ("Todo", "In Progress"))
+
     def test_label_routing(self):
         self.p.add_issue(1, labels=["model:astra"])
         self.p.add_issue(2, assignee="oregon", labels=["cap:gpu"])  # oregon lacks the capability