宜博 / lemongraph

The 6th real self-hosting attempt (first to reach the full 4-node doc-writing
graph after the previous five bugs) hit a genuinely new failure: this real
graph_manifest had no edges at all, not just wrong field names -- worse than
the case _normalize_graph_manifest already handled. With zero dependency
info to normalize from, topological_order() scheduled 'benchmark' before
'sdd' (alphabetically first among zero-in-degree nodes), and build_prompt()
crashed with KeyError: 'sdd_content' trying to format a template needing
sdd's not-yet-produced output.

_PROMPT_TEMPLATE always asks for exactly the same fixed 4-node linear chain
-- that's a constant architectural fact, not something that should be
re-inferred from the LLM's output every call. _normalize_graph_manifest now
deterministically enforces brd->prd->sdd->benchmark whenever all four
canonical artifact_types are present, regardless of what depends_on/edges
the LLM did or didn't provide. Same "机制用代码" principle as
ensure_worktree_committed and the earlier normalize logic.

Three new regression tests: zero dependency info still gets the chain
enforced, node_id != artifact_type still resolves to the real node_id (not
a guessed string), and a partial set (missing one of the four types) is
correctly left alone rather than guessed at.

560 tests green (440 src + 120 supervisor).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

The 4th real self-hosting attempt (first to reach the actual doc-writing
Work Graph after the previous three bugs) hit a real git merge failure:
"local changes would be overwritten by merge", exit code 2.

Root cause: _check_prd_consistency() calls execute_work_graph() directly
against target_folder_path, completely bypassing work_node_run.run_work_node's
worktree isolation. It uses agentType='planner' with memory:project, which
writes a real update to .claude/agent-memory/planner/MEMORY.md in whatever
cwd it runs with. Every prior real run of this check only ever targeted a
throwaway clone, so this never mattered -- until self-targeting, where cwd
IS the main checkout: the memory write landed as an uncommitted local
change on master, which then blocked the SDD node's own (correctly
isolated) worktree merge right after it.

Fix: apply 05 §5.8's same unconditional self-target worktree rule here,
reusing work_node_run's existing create_worktree/ensure_worktree_committed/
merge_worktree/remove_worktree rather than inventing a new mechanism.
Non-self-target calls are unaffected (verified by both regression tests).

Also cleaned up the real, now-recovered state from the failed 4th attempt:
merged a real SDD-node worktree branch that got stranded uncommitted
(conflict was just two near-identical memory file wordings, resolved by
hand), and removed three orphaned worktrees left by earlier test runs.

557 tests green (437 src + 120 supervisor).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

A real self-hosting run (U-042, first real exercise of needs_worktree=True
since U-035 forces it unconditionally for self-target) exposed a real bug
RealGitWorktreeIsolationTest never caught: its own run_fn always committed
its edits, but a real Coder isn't instructed anywhere to git commit its own
changes. The Coder genuinely edited src/org_loop_route.py and verified it
in its own test_report -- but never committed, so merge_worktree()'s `git
merge` was a true no-op ("Already up to date"), and remove_worktree() then
silently discarded the dirty worktree along with the real work. code_diff
came out empty; the change never reached the branch at all.

Adds ensure_worktree_committed(): checks `git status --porcelain` in the
worktree before merging and auto-commits any dirty state found. A no-op
when the worktree is legitimately clean (e.g. a pure investigation task).
Same "判断用 Agent，机制用代码" principle already used elsewhere in this
codebase -- don't rely on persona prompt-engineering for a mechanical
correctness guarantee that the system itself should hold.

Two new regression tests: one reproducing the exact "edited but never
committed" scenario, one confirming a legitimately-clean worktree still
merges fine. 554 tests green (435 src + 119 supervisor).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Adds src/deployment_query.py (list_deployments/get_current_deployment/
get_deployment -- no project_id filter needed, deployment has no project_id
column since there's exactly one graphlemon self-target) + three new
backend_server.py routes (GET /api/deployments, /current, /:id, the last
with recent deploy_health_signal history) + a new v_deployment_status page
in frontend/app.js polling every 5s (deployment.status changes don't go
through c_ws_subscribe -- Deploy Supervisor writes deployment/
deploy_health_signal directly, no execution_event of its own).

No 02-V spec exists for this view (PH5 was added after 02-V was written) --
DOM/class conventions follow the existing pages' own patterns. Verified
against a real running server via urllib (browser extension unavailable
this session), plus 9 deployment_query unit tests and 5 real HTTP route
tests.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Adds supervisor/deploy_health_check.py: run_health_check() runs the three
pre-cutover checks in order, stopping at the first failure (03-C: "任一检查项
fail 立即整体判定 health_check_failed，不等待其余检查项跑完") and persisting
a deploy_health_signal row for every check that actually ran.

- process_alive: polls deploy_supervisor.is_alive() up to HEALTH_CHECK_STARTUP_TIMEOUT.
- http_health: hits the candidate's /api/health -- added to backend_server.py
  (U-037 deliberately left this for U-038). Not just "process exists" --
  runs a real SELECT 1 to catch a wedged sqlite connection a live-but-stuck
  process would otherwise pass.
- canary_task: a real (not mocked) project+task creation round-trip against
  the candidate over HTTP -- no real Org Graph/claude calls, since
  c_task_create's synchronous 'queued' landing is already a deterministic,
  always-should-pass result, which is what a canary needs.

Timeouts are the MVP-fixed drafts from 05 §5.9, explicitly flagged there as
needing real calibration later, not hardcoded as authoritative.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Adds supervisor/deploy_stage.py: stage_deployment() creates a deployment row,
git-worktree-checks-out the target commit into a fresh directory (--detach,
never in-place git pull), copies the production DB to a one-time candidate
copy, and Popens a real candidate backend process on a free port.

Return value follows 04-P §4.6's state-machine table (status after the call
is 'candidate_starting') rather than 03-C's one-line 'staged' summary --
04-P's own trigger text ("c_deploy_supervisor_stage 完成" moves the row to
candidate_starting) is the more detailed/authoritative source when the two
disagree, same precedent used throughout this project.

Found and closed two real gaps along the way:
- deployment's own field list never included where a candidate's checkout
  directory lives -- added candidate_instance_dir to schema.py (also what
  the NEXT staging call's previous_instance_dir bootstraps from once a
  deployment reaches 'finalized').
- backend_server.py had no CLI entry point at all since U-007 -- create_server()
  was only ever instantiated in-process by tests. Added main() (argparse +
  serve_forever + SIGTERM->shutdown() from a separate thread, per
  http.server's own same-thread-deadlock warning).

RealCandidateLaunchTest spawns and HTTP-probes a genuine
`python3 -m src.backend_server` subprocess against a real git-worktree
checkout of this repo (not mocked) -- caught a real ordering issue doing so:
git worktree only sees committed content, so testing before this Unit's own
commit landed would check out backend_server.py without main() at all.
Fixed by using `git stash create` (builds a commit object from the current
working tree without touching HEAD/index) as the commit-to-test-against,
falling back to plain HEAD once the tree is clean post-commit.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Adds paths.repo_root()/is_self_target() and forces needs_worktree=True in
work_graph_orchestrate.run_graph() whenever target_folder_path is graphlemon's
own checkout, regardless of caller intent -- 05 §5.8's unconditional rule,
since a self-target in-place edit risks corrupting the currently-running
process's own source out from under it.

Adds src/core_engine_classify.py (path-based rule: .claude/agents/*.md,
src/*, supervisor/* = core_engine; frontend/, docs/ = not) and
bug_record.record_core_engine_change() (a PASS-path audit record, distinct
from create_bug_record's FAIL-path root-cause attribution -- no route
decision, status='resolved' from creation). scheduler.finish_task_lifecycle()
records this audit trail as an additive side effect after close_task_done()
on self-target PASS closures.

Found and resolved a real SDD-internal conflict (confirmed with the user via
AskUserQuestion) before wiring this up: supervisor-auto-rollback-design.md §1
triggers Deploy Supervisor on self-target+Validator-PASS alone, but 03-C
§3.11/04-P §4.6 literally gate it on an existing core_engine bug_record --
and task_create.py confirms goal-type self-improvement tasks (the exact
self-development-feasibility.md walkthrough scenario) never produce a
bug_record on their happy path, which would silently defeat that whole task
class under the literal reading. Resolved per user decision: trigger on
self-target+PASS alone; the bug_record is a byproduct, not a gate.
p_deploy_supervisor_blue_green (04-P §4.6) never touches task.status, so
close_task_done() still runs unconditionally -- Deploy Supervisor is a fully
decoupled state machine, matching §2's "no shared failure domain" principle.

org_loop_route.py is intentionally untouched -- fast_lane/full_replan is
FAIL-path retry routing, orthogonal to this PASS-path trigger; deployment
row creation stays U-037's job per its own Done criterion.

417 tests green; RealGitWorktreeIsolationTest (U-013/REQ-03's real git
worktree regression) re-verified passing per this Unit's own Done criterion.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Adds the two Deploy Supervisor entities from 01-M §1.15/§1.16 (17 tables
total): deployment (full 15-value blue-green state machine, per 01-M's own
text rather than supervisor-auto-rollback-design.md §3's abbreviated
summary table) and deploy_health_signal (append-only health/monitoring
samples). ux_deployment_active enforces 01-M's "only one non-terminal
deployment at a time" constraint via an expression-based unique index
(CREATE UNIQUE INDEX ... ON deployment((1)) WHERE status NOT IN (...)) --
the same partial-unique-index idiom work_graph_run already uses, just with
a constant instead of a real grouping column since there's exactly one
graphlemon self-target.

bug_record.root_cause_category's schema-level CHECK now accepts
core_engine, but bug_record.py's own application-layer whitelist
deliberately does not follow yet -- classifying a bug as core_engine is a
file-path-based rule (05 §5.8), not something the Validator's existing
3-perspective judge_root_cause can produce, and wiring that classification
+ c_org_loop_route's routing is U-035's job.

392 tests green.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

CLAUDE.md is explicit that PH5 (self-modification/auto-rollback) must not
start until PH4 has genuinely closed -- not just "all unit tests green,"
but 06-验收与测试锚点.md's REQ-01~14 happy/boundary/failure cases actually
covered. A REQ-by-REQ audit (the same discipline that caught U-031's
integration gap) found 5 real gaps:

- REQ-03 failure: work_node_execution.status jumped straight
  running->retrying, skipping 'reflecting' (declared in the schema enum,
  never once written by any code).
- REQ-05 failure / REQ-09 boundary: the "scheduler keeps serving other
  queued tasks past one blocked_pending_human" scenario was never tested
  against a real scheduler -- the old test literally said "no scheduler
  exists yet to test against directly" (stale since U-032).
- REQ-06 / REQ-07 / REQ-14 failure (the biggest gap, shared root cause):
  CorruptPersistedStateError, ClassifierBlockedError, and
  UsageLimitWaitTimeoutError were each correctly raised and each correctly
  marked their own work_node_execution 'blocked', but nothing ever routed
  that to task.status='blocked_pending_human' with the blocked_reason each
  REQ's acceptance text names. e2e_pipeline.run_end_to_end is split into
  run_task_lifecycle() so scheduler.py's run_goal_to_completion can create
  the task itself and catch these three specific exceptions, following the
  same UPDATE+execution_event pattern budget_guardrail.check_budget()
  already established for REQ-10.
- REQ-02 boundary: PRD internal-contradiction check (verifying->reflecting)
  had no implementation anywhere -- artifact_write always marked
  validation_status='passed' unconditionally. Added an optional verify_fn
  hook to work_graph_orchestrate.run_graph() (one regeneration attempt,
  then ConsistencyCheckFailedError) and a real consistency-check agent()
  call scoped to the PRD node only. The first real run of this new call
  caught a genuine bug: the model returned "consistent": "false" as a JSON
  string, not a boolean, which `if ok:` silently treated as truthy -- fixed
  with explicit _coerce_bool().
- REQ-13 failure: minor test-assertion weakness (WS 404 rejection test
  didn't check the JSON error body).

384 tests green. REQ-01~14 now have real, exercised coverage rather than a
"tests are green" assumption -- PH4 can be treated as genuinely closed.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

04-P's blocked_pending_human branch, 01-M's task-table indexing rationale,
and REQ-09's boundary case all presume an autonomous dispatcher scans queued
tasks and keeps serving the queue past a blocked one -- but no such component
existed. org_node_lifecycle._check_concurrency (U-010) only gates a single
trigger call; e2e_pipeline.run_end_to_end (U-014) drives exactly one task,
synchronously, only ever called from tests.

Adds src/scheduler.py: run_scheduler_once() dispatches multiple Goals via
ThreadPoolExecutor bounded by org_concurrency_config.max_concurrent_goals,
each worker opening its own sqlite connection (with a busy_timeout) rather
than sharing one across threads -- U-027's SIGSEGV already proved that needs
a hand-held lock, which would just serialize everything back to front.

Also closes a gap run_end_to_end's own docstring already flagged: it never
called close_task_done (PASS) or org_level_loop.run_fix_loop (FAIL), leaving
every task stuck at 'validating'. finish_task_lifecycle() applies that step
without changing run_end_to_end's own tested behavior or scope.

Real-write concurrency (multiple threads, separate connections, same db
file) is stress-tested for real in test_scheduler.py without spending API
budget -- the actually-new risk surface here is the threading/connection
design, not the already-validated (U-031/U-017) work itself.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

U-031: closes a structural integration gap found during the REQ-01~14 regression
audit -- plan_work_graph/execute_work_graph/run_work_node were each individually
tested but never actually called together in production code (e2e_pipeline.py's
only validated closed loop bypassed all three via flat org_node_trigger calls).
Adds work_graph_orchestrate.py (topological sort over graph_manifest + one
one-node dynamic-workflow agent() call per Work Node, needed for REQ-06's
per-node breakpoint recovery), rewires e2e_pipeline.py's plan_documents/
implement_code onto the real chain, and fixes two bugs a real API run
surfaced: graph_manifest's id/node_id/depends_on normalization
(work_graph_plan._normalize_graph_manifest) and a missing repo_path in
run_work_node's cwd resolution. work_node_run.py also now marks a node
'blocked' instead of leaving it stuck 'running' on any unretryable run_fn
exception.

U-024.5: the regression audit also found REQ-12's reconnect boundary case
(refetch a snapshot after a dropped WS reconnects) was never implemented --
only the dashboard reacted to onStatusChange, and even it never refetched.
Fixes mountDashboard/mountTaskBoard/mountTaskDetail (which embeds
v_graph_visualizer) to refetch their current snapshot on reconnect.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

三个都是 config_id='default' 单例配置，首次使用即创建、之后原地更新
（upsert），不需要单独的初始化流程：

- c_concurrency_config_update：新增 src/settings_config.py（
  org_concurrency_config 之前完全没人碰过），校验 max_concurrent_goals
  落在 1-50。
- c_budget_config_update：budget_guardrail.py 新增 upsert_budget_config()
  ——跟 U-019 已有的 update_budget_limit_and_resume()（只更新已存在的
  行）不是一回事，这个是真正完整的设置页控制器，支持首次创建，沿用
  同一套"额度调高后自动恢复 budget_exceeded 阻塞任务"副作用（提取成
  _resume_budget_blocked_tasks() 共享）。
- c_log_retention_config_update：同样在 settings_config.py 里，校验
  retention_days > 0。

三处校验都在业务层做完整检查，HTTP 层额外兜了 except (InvalidParamError,
TypeError)——payload 数值字段缺失时 None<=0 这类比较会抛 TypeError 而
不是业务异常，两个都要接住才能保证返回干净的 400 而不是裸 500。
backend_server.py 第一次加了 do_PUT。

前端三个页面共享一个 renderSettingsSubnav() 小组件模拟 02-V 原型里的
设置下拉菜单。真实服务器端到端验证过三个 PUT/GET 的完整读写循环。

PH3（人机交互控制台）至此全部完成：U-022~U-030。

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

c_skill_diff 和 U-028 的 c_artifact_get 一样，只存在于机器可读的
sdd-mvcp-contract.json，03-C 正文从未写出来，已订正补全。
c_skill_list/c_skill_get 两份文档都没定义，是 v_skill_library/
v_skill_audit_log 自己的展示需求倒逼出来的缺口。

新增 src/skill_query.py：list_skills()、get_skill_audit_log()、
get_diff()（git diff <before> <after>——skill_update.py 的
_write_and_commit 把每次迭代做成原子单文件 commit，整仓库范围的 diff
天然只涉及那一个 Skill 文件）。

关键安全考量：Skill 库活在 graphlemon 自己的仓库里，c_skill_rollback
会真的在这个仓库里跑 git commit——如果 HTTP 层写死指向真实仓库根目
录，任何针对这条路由的测试都会在正在跑的这个 graphlemon 仓库里留下
真实提交。参照 db_path 已有的做法，给 create_server() 加了
skill_repo_path 参数（默认 graphlemon 真实仓库根，测试传临时仓库），
不通过 HTTP 暴露这个参数，避免变成任意写入的口子。记录在 05
§5.4.1.4。

真实服务器 + 临时仓库端到端验证过完整的 update→diff→rollback 闭环，
确认过跑完之后 graphlemon 自己仓库的 git log 完全没变化。

v_skill_library/v_skill_audit_log 不检查"当前项目"——Skill 是平台级
资产，跟项目无关。

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

c_artifact_get 在机器可读的 sdd-mvcp-contract.json 里其实定义过
（GET /api/artifacts/:artifact_id），只是 03-C 正文一直没写出来，已订正
补全。c_artifact_list 连 MVCP JSON 里都没有 operation 定义，纯粹是
v_artifact_browser 自己的过滤字段倒逼出来的缺口，跟 U-025 补 c_task_list
是同一类问题。

新增 src/artifact_query.py：list_artifacts() 返回裸数组（跟已有的
list_tasks() 保持同款风格）；get_artifact() 把血缘链（按
depends_on_artifact_ids 往回走，只走第一个依赖——这条流水线的实际产出
只有线性 BRD→PRD→SDD 链）和文件内容（artifact 表本身没有 content 列，
走 project.target_folder_path + artifact.file_path 现读磁盘）一起返回，
预览面板一次请求就够。

前端预览统一用 <pre> 纯文本展示，不做 Markdown 渲染/代码高亮——02-V
原文想要那个效果，但需要引入第一个前端第三方库，跟这个控制台一贯的纯
vanilla JS、无构建步骤做法冲突，是有意简化。顺手给 index.html 常驻导航
栏加上了"产物"链接。

真实服务器验证过三级血缘链（BRD→PRD→SDD）返回顺序正确。

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

execution_log_file 的行生命周期 U-013 早就管好了，但从来没有代码真的往
日志文件里写过内容——src/execution_log.py 新增 append_log() 补上这条
缺失的写路径，设计成 run_fn 执行过程中随时可调用，不关心节点在做什么。
get_log() 按 offset/limit 分页读，download=true 时走原始文件流。

c_ws_log_tail（WS /api/ws/work-nodes/:node_exec_id/log）复用 U-024 手写
的 RFC 6455 机制，轮询 execution_event(log_appended)；节点一旦转终态
（done/blocked/failed）服务端主动发关闭帧断开，不让客户端一直轮询。

p_log_retention_cleanup 实现成纯函数 run_log_retention_cleanup()：过期
扫描 + max_total_size_mb 超限按最旧优先追加候选 + 单个文件删除失败只
记录不阻塞其余项，没有配调度器（不在本 Unit 覆盖范围），留给未来的
调度进程周期调用。

过程中意外抓到一个真的并发 bug：_Server.server_close() 关闭共享
sqlite 连接时原来没拿 conn_lock，如果这时候正好有个 WS 轮询线程在
另一边持锁执行 SQL，两个线程无同步地并发操作同一个 C 级 sqlite3 连接
对象，实测直接把解释器崩了（SIGSEGV）——U-024 就已经有这个隐患，这次
高频 WS 测试把时序窗口撞开了。修好后连续跑几十次全部稳定通过，记录
在 05 §5.4.1.3。

frontend/app.js 的 v_log_viewer 挂在 v_task_detail 每个 to-do 节点的
展开详情里，首次展开才建连接（懒加载），运行中态走 c_ws_log_tail 实时
滚动、终态走 c_execution_log_get 静态拉取，活跃连接在任务级刷新/离开
页面时统一关闭。真实服务器 + Node 原生 WebSocket/fetch 端到端验证过
实时追加与终态自动断开的完整链路。

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

纯前端 Unit，数据已经在 U-025 的 c_task_get 里，只是 task_query.py
的 _work_nodes_for_run 补上了 hover tooltip 需要但当时没取的字段
（input_schema_ref/output_schema_ref/quota_limit_type/
quota_wait_started_at）。

frontend/app.js 新增 computeGraphLayers()：按 graph_manifest 节点的
depends_on 边做拓扑分层，渲染成按执行层级排布的行，fan-out/fan-in
都能正确分层——用一个 n1→[n2,n3]→n4 的真实 graph_manifest 验证过，
直接从 app.js 里取出 computeGraphLayers 本体（不是重新实现一份逻辑）
跑真实服务器返回的 JSON，分层结果完全正确。节点按 03-C 3.6 节六态
配色，waiting_for_quota 态 hover 展示 quota_limit_type + 已等待时长。

02-V §2.7 原描述还包含一份 Org Graph 层，但 v_task_detail 页面顶部
已有等价的 org-strip 面板，这里判断重复渲染没有意义，只做了 Work
Graph DAG 层，Org Graph 复用页面已有的那个。

嵌入位置：v_task_detail 里原来"随 U-026 补上"的占位提示换成了真实
的 DAG 面板。

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

三个页面需要的读接口此前都不存在，本 Unit 一并建了：

- c_task_list/c_task_get（03-C §3.2，此前无 Unit 认领）：新增
  src/task_query.py + GET /api/tasks、GET /api/tasks/:task_id。
  c_task_get 的返回比 03-C 字面契约更完整——额外带上
  work_graph_run（含解析后的 graph_manifest）+ 每个 run 的
  work_node_execution 列表，因为 v_task_detail 的 to-do 树需要这份数据，
  03-C 已订正补全。
- c_org_loop_route（U-016 已实现 route_org_loop()，从未接 HTTP）第一次
  接上 POST /api/tasks/:task_id/route，v_task_detail 的人工干预
  （manual_resume）按钮需要它。
- 预算护栏此前没有任何读接口（03-C 只写了 PUT），新增
  budget_guardrail.list_budget_configs() + GET /api/settings/budget
  给仪表盘预算条用。

frontend/app.js 补上真实的 #/（org graph 三节点卡片、当前任务概要、
预算条、事件时间线，全部通过 c_ws_subscribe 实时刷新）、#/tasks
（关键字/状态/org_node 过滤）、#/tasks/:task_id（task 字段、to-do 树、
关联 artifact/bug_record、blocked_pending_human 告警条 + manual_resume
按钮）。路由改成对象形式支持挂载态清理，切页面时关掉上一个页面的
WebSocket 连接。v_graph_visualizer/v_log_viewer 子模块明确标成占位
提示（随 U-026/U-027 补上），不是假造内容。

顺带把 U-023 里因 c_task_list 当时不存在而简化的 related_task_id
换回真下拉，goal_composer/bug_feedback 提交成功后也改回跳转真实的
v_task_detail。

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Python stdlib 没有 WebSocket 支持，src/ws_subscribe.py 手写 RFC 6455
握手（hashlib+base64 算 Sec-WebSocket-Accept）+ 帧编解码（服务端帧不
加掩码，客户端帧强制解掩码），不引入 websockets 之类第三方库
（05 §5.7）。

推送用轮询而非写路径 hook：execution_event 的写入分散在十几个调用点
（org_node_trigger/work_graph_execute/work_node_run/task_create/
bug_record/...），没有单一写入路径可以插桩，backend_server.py 的
_handle_ws_subscribe 劫持 ThreadingHTTPServer 连接后每 0.5 秒按
rowid 增量查一次 execution_event（project_id/可选 task_id 过滤），
轻松满足 REQ-12 的 ≤2 秒目标。断线重连按 03-C §3.14 走 REST 快照，不
在 WS 层做消息重放，新连接只看连接那一刻起的新事件。

daemon_threads = True 避免长连接 WS 线程阻塞进程退出。

验证：test_backend_server.py 用原始 socket 走真实握手+推送延迟断言；
另外用 Node 原生 WebSocket 客户端做跨实现互操作性验证，连真实运行的
服务器，从建连到收到 c_task_create 触发的真实推送全程走通。

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

c_task_create（U-009 已实现的 task_create.create_task()）第一次接上
POST /api/tasks。c_bug_report_create 此前完全没有实现，新增
bug_record.create_bug_report() + POST /api/bug-reports。

发现并订正 03-C 一处输入契约遗漏：c_bug_report_create 的
input_contract 没写 project_id，但 execution_event.project_id 是
NOT NULL、bug_record 表自己没有 project_id 列可以在 related_task_id
留空时兜底推导，字面上无法满足约束——按 c_task_create 同款处理方式把
project_id 变成必填，03-C 已订正。

frontend/app.js 补上 #/goals/new、#/bugs/new 真实交互；控制台还没有
"当前项目"状态的正式载体（v_dashboard 是 U-025 才做），project_id 暂存
localStorage，v_project_setup 创建成功时写入。related_task_id 因为
c_task_list 未被任何 Unit 认领，简化成文本输入而非下拉，等 U-025 补上
任务列表后再换。提交成功后原型写的是跳转 v_task_detail，该页面也是
U-025 才建，暂时改成原页面内展示确认信息。

端到端冒烟时额外发现一个真 bug：create_bug_report() 原本没校验
project_id/related_task_id 是否存在，坏 project_id 会在 execution_event
插入时触发未捕获的 IntegrityError，请求直接 500——补了存在性校验，
改为干净的 INVALID_PARAM/400。

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

c_pick_folder 补上 Windows（PowerShell FolderBrowserDialog）和 Linux
（zenity 优先、回退 kdialog）实现，三平台都覆盖到。c_project_create
（U-002 已实现的 project_scaffold.create_project()）第一次接上 HTTP
路由 POST /api/projects——backend_server.py 因此第一次需要持有 sqlite
连接，schema.connect() 加 check_same_thread=False 参数，配合
server.conn_lock 序列化 ThreadingHTTPServer 各请求线程的访问。
frontend/app.js 补上 #/setup 真实交互（选文件夹→校验 Git 仓库→非
Git 仓库二次确认→创建→跳转仪表盘），DOM class 命名对齐
docs/wireframes/v_project_setup.html。

REQ-11 完整 happy 用例（含创建 task）还需要 U-023 的
v_goal_composer/c_task_create，本 Unit 只完成 v_project_setup/
c_pick_folder/c_project_create 部分，development-plan.md 已记录。

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

now 默认值从 time.monotonic 改成 time.time（墙钟），使 quota 等待的
已耗时计算能在进程重启后正确续算；run_work_node 对已 done 的节点直接
返回存量结果不重跑，对 retrying/waiting_for_quota 节点从持久化的
retry_count/quota_wait_started_at 续算而非清零；needs_worktree=True
时复用已存在的 worktree 而不是重新 create。新增
resume_work_graph_run() 返回一次 run 里尚未完成的 node_exec_id 列表，
并对内部不一致的持久化状态统一抛 CorruptPersistedStateError。

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Re-checking REQ-14's boundary case against the letter of its text found a
real functional gap, not just a testing gap: U-013's opus handling only
did "skip backoff" -- it never actually retried on a downgraded model,
just failed immediately. REQ-14 boundary says "不走退避轮询，直接降级到非
Opus 模型重试" (skip backoff, retry immediately on a non-opus model) --
the retry was the part that was missing.

Fixed: run_fn's signature is now run_fn(cwd, model=None); an opus quota
signal retries immediately (no backoff) with opus_downgrade_model
(default "claude-sonnet-5"). If the downgraded model somehow also reports
an opus signal (shouldn't happen -- it's not using Opus), this doesn't
loop forever chasing it, raises UsageLimitWaitTimeoutError instead.

work_graph_execute.py's own opus handling (U-012) deliberately NOT changed
to match: REQ-14's coverage only names c_work_node_run/c_org_node_trigger,
and that mechanism (a multi-agent dynamic workflow, no single top-level
--agent identity) has no one "the model" to downgrade -- a deeper A-1
question, documented as out of scope rather than silently left
inconsistent without explanation.

Also filled in a real test gap: quota_limit_type=weekly (detection +
recovery, and the 9-day timeout) was never tested despite the mechanism
already supporting it since U-013.

11 new/changed stdlib unittest cases across work_node_run.py's test suite
(all existing run_fn lambdas updated for the new signature). 178 total
across U-001-020, all passing.

This closes PH2-S3 alongside U-019. REQ-14 is now genuinely complete, not
just "mostly right."

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Found running U-014's full real closed loop: the Validator gave an
INCONCLUSIVE verdict and explicitly said it had no Bash tool, despite its
frontmatter granting one. Isolated with a minimal test subagent: any
"Bash(...)" scoped pattern (e.g. "Bash(git commit:*)") in a subagent's
disallowedTools frontmatter field disables Bash *entirely*, not just that
subcommand -- confirmed by comparing a disallowedTools=[Edit,Write] agent
(Bash works) against one adding a single Bash(git commit:*) entry (Bash
breaks completely). The CLI's own --disallowedTools flag documents this
exact scoped syntax as supported; subagent frontmatter just doesn't honor
it the same way, and nothing in the docs says so.

Fixed agent_profile.py's validator FRONTMATTER to disallowedTools=
[Edit, Write] only -- dropping the four Bash(git *) entries that were
silently breaking test execution. The "don't commit/push/reset/clean"
constraint now relies on the prose instruction already in the migrated
AGENTS.md instead of a tool-level block that never actually worked; fixed
that file's own (now-false) claim that it was enforced at the tool level,
and documented the finding centrally in 05-实施约束与依赖.md §5.4.1.2 since
it'll matter for any future agent definition, not just this one.

Verified for real, redeployed: Bash now works for the actual Validator
identity, Edit/Write remain correctly blocked.

1 test fixed (was asserting the now-removed scoped pattern) + 1 new test
(disallowedTools frontmatter must never contain "Bash(" again). 174 total
across U-001-019, all passing.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

check_budget(): accumulates current_usage on success (04-P monitoring->
monitoring), and on reaching/exceeding limit_value transitions the task to
blocked_pending_human with blocked_reason='budget_exceeded' (never
'usage_limit_wait_timeout' -- 04-P is explicit these two guardrails, human-
configured budget vs Claude Code's own plan quota, must not share a
blocked_reason) and raises BudgetExceededError rather than returning
allowed=false silently, matching 03-C §3.12's own errors list.

update_budget_limit_and_resume(): REQ-10's boundary case needs "raise the
limit -> paused tasks resume" to be real; built the minimal version this
Unit needs rather than the full settings-page controller
(c_budget_config_update, U-030/PH3's job) -- reuses U-016's manual_resume
(org_loop_route.route_org_loop) for the actual status restoration instead
of reimplementing it.

6 new stdlib unittest cases, all real (sqlite, no mocking needed). 173
total across U-001-019, all passing.

Also: the U-014 full real closed-loop run that's been going in a detached
background process since earlier this session finished -- 957s (~16min).
All 4 real BRD/PRD/SDD/Benchmark docs, a real JS snake-game implementation
(engine.js/engine.test.js/index.html/app.js) from the Coder, and a
genuinely careful INCONCLUSIVE verdict from the Validator: it explicitly
refused to accept the Coder's own claimed "10/10 passed, verified with
headless Chrome" without being able to reproduce it, correctly citing its
own migrated SOUL.md rule ("工具错误和权限不足属于INCONCLUSIVE，不能洗成PASS").
It also caught a real test-fidelity gap (T3/T7 not matching the Benchmark's
literal wording) via careful static tracing. Result preserved at
/private/tmp/.../scratchpad/snake-game-result for inspection. The Validator
also reported no Bash tool available despite its frontmatter granting one
-- investigating that as a follow-up, separate from this commit.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

update_skill(): writes new Skill content to skills/<skill_id>/SKILL.md and
commits it (03-C §3.8: file write + git commit must be atomic -- the DB
insert only happens after the commit succeeds, so a git failure naturally
leaves no orphaned skill_iteration row, no separate rollback machinery
needed), rejects a missing triggered_by_bug_id (MISSING_TRIGGER_REF).
rollback_skill(): restores the pre-iteration content from
before_git_commit_sha and commits *that* as a new commit + new
skill_iteration row (03-C §3.9: rollback is itself a new iteration, not a
history rewrite); the rollback record's triggered_by_bug_id reuses the
original iteration's, since c_skill_rollback's input contract has no bug_id
of its own and that FK is NOT NULL.

repo_path has no default (caught and fixed during this Unit, not shipped
with the bug): an earlier draft defaulted it to this repo's own root,
which would have made any test that forgot to override it silently commit
into graphlemon's actual git history. Every call site must now be
explicit; git log confirms nothing leaked into this repo while building
and testing it.

Resolves U-015's deferred skill_defect gap: re-reading 04-P §4.3's own
transition table, bug_recorded (c_bug_record_create alone) and
skill_iteration_triggered (-> c_skill_update) are separate, sequential
transitions, not one atomic call -- so 03-C §3.10's "必须同时创建
skill_iteration_suggestion" is a same-flow requirement, not a same-API-call
one. bug_record.py already accepted skill_defect (it was never actually
rejected, just undocumented as resolved); update_skill() now fills in
bug_record.skill_iteration_suggestion_id once the real iteration exists.

Known gap, not this Unit's job: no real Skill content has been migrated
from the Hermes profiles yet (05 §5.1 names the skills; agent_profile.py's
frontmatter only references them by name) -- tests seed a placeholder
skill file, same pattern as U-010/011 seeding placeholder artifacts for
upstream producers that don't exist yet.

10 new stdlib unittest cases, all real (git + sqlite, no LLM call needed --
same category as U-013's worktree tests). 167 total across U-001-018, all
passing.

This closes PH2-S2. REQ-08 is now real code, and U-015's skill_defect note
is resolved rather than left permanently deferred.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

run_fix_loop() chains U-015 (3-perspective root-cause judgment) + U-016
(routing) + real Coder/Planner/Validator calls into architecture doc §11's
4th MVP acceptance criterion: judge -> create bug_record -> route
(fast_lane for category=coder, full_replan for category=planning -- a
fixed mapping this Unit defines, since bug_record.suggested_route is
technically an independent Validator judgment but REQ-05/architecture §11's
scenarios always correlate the two, so no extra Agent call was added just
to re-decide it) -> trigger the fixer -> re-trigger Validator -> close_task_
done on confirmed PASS.

3 new mocked stdlib unittest cases (route branching, verification-fails
leaves the task out of 'done'). 160 total mocked tests across U-001-017,
all passing.

Real coverage reuses U-015's wall-collision scenario but this time drives
an actual fix: a real Coder call edits collision.py in the repo, a real
Validator call reads the fixed file and confirms it, running as a detached
background process (survives this session's own restarts, unlike two
earlier real runs this session that got silently orphaned by a session
teardown with no completion record -- lesson learned, not repeated).

This closes PH2-S1 (U-015/016/017): REQ-04/05/09 and architecture doc §11
criterion 4 are now real, verified code, not just planned intent.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

route_org_loop(): fast_lane -> coding/coder, full_replan ->
planning_brd_prd_sdd/planner, loop_round += 1, writes an
execution_event(event_type=loop_routed) recording the decision.
loop_round >= 3 forces blocked_pending_human (blocked_reason=
'loop_limit_reached') and raises LoopLimitReachedError instead of routing
(03-C §3.11: "不接受常规路由请求，只接受 manual_resume"). manual_resume walks
execution_event's task_status_changed history backwards to find what
status preceded the most recent transition into blocked_pending_human, and
restores it -- 04-P says "恢复到触发前状态" but task has no
previous_status column, so this reconstructs it from the event log
already being written everywhere in this codebase, rather than adding a
new column for a value that's already reconstructable.

Also built close_task_done() (validating->done, validator_pass): 04-P
names this transition but not a controller for it, and nothing in this
codebase had ever set task.status='done' before this commit -- REQ-05's
happy case ends there, so it had to exist for that to be reachable.
Documented as a deliberate scope extension, not silently assumed.

Simplification, noted rather than silently collapsed: 04-P models
fast_lane/full_replan as two hops (validating -> routed_fast_lane ->
coding) but c_org_loop_route's own 03-C contract returns the final state
directly and no controller owns the second hop -- implemented as one
atomic transition, with the route decision preserved via the loop_routed
event instead of a persisted intermediate status. Neither REQ-05 nor
REQ-09 check for that literal intermediate state.

11 new stdlib unittest cases, all real (sqlite is fast/free/local, no
mocking needed). 157 total across U-001-016, all passing.

This closes PH2-S1 (U-015 + U-016): REQ-04, REQ-05, and REQ-09 are now
all real code, not just planned.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

judge_root_cause(): runs 3 perspective_fns (correctness/requirement-
completeness/skill-defect, 04-P §4.3), requires >=2/3 agreement on a
category or raises InconclusiveEscalateError with all raw judgments
attached (REQ-04's failure case). trace_root_cause_test_failed() creates a
brand-new bug_record on agreement; trace_root_cause_human_reported()
field-completes an existing one instead (04-P's explicit distinction
between the two origins, not a new record either way).

c_bug_record_create (bug_record.py): validates root_cause_category against
{planning, coder, skill_defect} -- core_engine deliberately excluded,
matching U-001's schema CHECK constraint (PH5/U-034 scope). Known,
deliberately-unenforced gap noted rather than guessed at: 03-C §3.10's
"skill_defect requires creating a skill_iteration_suggestion" constraint
can't be satisfied yet (skill_iteration.before/after_git_commit_sha are
NOT NULL but don't exist until a real skill rewrite happens in U-018) --
REQ-04's three cases never exercise skill_defect, so nothing here needs it
resolved yet.

real_validator_perspective() builds a real claude --agent validator call
framed from one of the three perspectives, parsing a CATEGORY:/DETAIL:
response format.

Also fixes a real diagnostic bug found investigating an unrelated failure:
org_node_trigger.py/work_graph_plan.py/work_graph_execute.py's nonzero-exit
error paths only included stderr in the raised exception, silently
dropping stdout -- which is where a real `claude` failure's actual detail
often lives. A real Coder call failed with exit 1 and empty stderr,
giving zero diagnostic signal; fixed to include both streams (truncated).

18 new mocked stdlib unittest cases + 1 gated real test (3 real Validator
perspective calls judging a deliberately-injected wall-collision bug,
expecting agreement on root_cause_category=coder). 146 total mocked tests
across U-001-015, all passing.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Closes the c_artifact_write gap U-011 flagged (03-C §3.7, unowned by any
Unit despite REQ-02 needing it directly): src/artifact_write.py writes
content to the target project, commits it to the target's own git history,
validates depends_on_artifact_ids against existing rows before writing
anything (REQ-02's failure case), creates the immutable artifact row.
validation_status is set to 'passed' unconditionally -- noted as a real
simplification (semantic content review, e.g. REQ-02's "PRD 里出现矛盾" boundary
case, needs its own LLM-judgment step this codebase doesn't build yet).

src/e2e_pipeline.py orchestrates architecture doc §11's MVP criteria 1-3:
plan_documents() makes 4 real sequential Planner calls (BRD, then PRD
referencing BRD, then SDD referencing PRD, then Benchmark referencing SDD),
writing each via c_artifact_write with correct lineage. implement_code()
calls Coder once, captures the resulting `git diff` as code_diff and the
Coder's own response as test_report. validate_implementation() calls
Validator once and parses an explicit 'VERDICT: PASS|FAIL|INCONCLUSIVE'
line. run_end_to_end() chains create_task -> plan -> implement -> validate,
using U-010's trigger_org_node_full only for task.status bookkeeping (a
lightweight "confirm ready" call per phase) and plain org_node_trigger
calls for actual content generation, since trigger_org_node_full's
contract returns a status dict, not text.

Real-verified: plan_documents ran for real against "实现贪吃蛇游戏" -- all 4
documents produced, correctly lineaged, validation_status=passed, real
content (>100 chars each), ~5m45s total for the 4 sequential calls.
Deliberately NOT run for real: run_end_to_end's Coder/Validator half (real
code generation + real test execution) is a different order of magnitude
in cost/time/uncertainty than every other real call this session -- built
and mock-tested, but not executed without checking first.

19 new mocked stdlib unittest cases (4 for artifact_write, 5 for the
pipeline orchestration, all real/unmocked for artifact_write since git
operations are fast/free/local same as U-013's worktree tests) + 1 gated
real test. 127 total mocked/real-local tests across U-001-014, all passing.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

run_work_node(): the per-node contract -- input schema check (rejects
outright, never starts, vs. output schema failure which retries), status
machine (pending->running->(reflecting/retrying up to caller-specified
max_retries)->done|blocked), execution_log_file lifecycle (active on start,
closed in a finally so it closes even when run_fn raises), and
USAGE_LIMIT_REACHED handling using 03-C §3.6's *exact* detection regex
(`hit your (session|weekly|opus) limit`, case-insensitive) -- more precise
than the phrase-matching approximations U-010/U-012 used before this Unit
existed to give quota_wait_started_at/quota_limit_type a real owner.

git worktree isolation (create_worktree/merge_worktree/remove_worktree)
tested for real, no LLM call needed: two independent nodes each get their
own worktree + branch, each run_fn writes and commits a different file,
both merge back into the main checkout with --no-ff, git log confirms both
commits landed, git_worktree_ref differs per node. This is REQ-03's
boundary case, verified end to end with real git.

"actual work" is an explicit required run_fn(cwd) dependency, not a hidden
default calling `claude` -- what a node does is inherently node-specific
(the whole point of "give each node a contract"); production callers wire
in a real subprocess call, tests inject fakes.

Corrects the same Done-line overreach as U-009/010/011/012 (5th occurrence):
REQ-03's happy case needs c_artifact_write (still unowned, per U-011's
flagged gap) and a real Coder session, neither of which U-013 has access
to via its own dependencies (U-012 only). Boundary is what's real here.

10 new stdlib unittest cases, all real (no mocking needed -- git and the DB
are both fast/free/local). 117 total across U-001-013, all passing.

This closes PH1-S2, the sub-phase development-plan.md's own text flagged as
"本项目风险最高的 Sub-phase" (highest-risk in the project).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Extends U-006's execute_work_graph in place (same controller, being
completed, not a parallel module):
- ClassifierBlockedError: raised when a response's permission_denials array
  has >=3 entries (03-C §3.5 says "3 consecutive or 20 cumulative" -- true
  cross-call cumulative tracking needs persistence U-013 doesn't provide
  yet, so this is a documented, conservative per-call approximation).
- UsageLimitReachedError (carries quota_limit_type) + a new
  execute_work_graph_with_usage_limit_retry() implementing p_usage_limit_wait
  (04-P §4.4.1): 5->10->20->30min backoff (repeating at 30) for
  session/weekly quota hits, timeout after 8h/9d -> UsageLimitWaitTimeoutError;
  opus quota hits skip backoff and fail immediately (model-downgrade retry
  is a caller decision, not this function's). sleep/now are injectable so
  tests don't block on real waits.

Retry operates at the whole-work-graph-call level, not per individual
work_node_execution row -- U-013 (c_work_node_run, which actually owns that
table) doesn't exist yet, so node-level quota tracking isn't attempted
here; noted explicitly in development-plan.md rather than silently
simplified.

Also corrects the same class of Done-line overreach as U-009/U-010/U-011
(4th occurrence): REQ-07's happy case literally asks for REQ-01~06 all
passing end to end, which needs PH2 (Validator root-cause tracing, Org-level
Loop, breakpoint recovery) -- U-012 only owns REQ-07's boundary/failure and
REQ-14's happy case, which was already correctly scoped.

23 new/changed mocked stdlib unittest cases (existing U-006 tests untouched
and still passing -- subclassing WorkGraphExecuteError keeps them
backward-compatible). 107 total mocked tests across U-001-012, all passing;
both existing real-agent tests (U-006's reused spike, U-011's planner call)
re-verified against the refactored code.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

plan_work_graph(): real `claude --agent <org_node> -p` call, produces a
graph_manifest (4 nodes for brd/prd/sdd/benchmark, dependency-chained
sdd->prd->brd, each edge tagged zero_token/needs_judgment per 03-C §3.4),
recorded as a work_graph_run row (status='planning' -- 04-P
p_work_graph_execution start_state). Does not execute the plan or produce
real artifacts -- 03-C §3.4 says so explicitly ("不执行，只规划").

Real call surfaced a genuine, undocumented constraint: --json-schema is
silently ignored when combined with --agent (structured_output never
appears; the identity itself works fine -- the response referenced the
migrated SOUL.md's own principle numbering). First attempt at this Unit
failed against the real API for exactly this reason. Fixed by asking the
model to fence its JSON and parsing it out of `result` instead
(_extract_graph_manifest, tolerant of bare JSON / fenced blocks / an extra
wrapper key). Documented centrally in 05-实施约束与依赖.md §5.4.1.1 since
this will matter for any future Unit needing agent-identity + structured
output together, not just this one.

Also corrects the same class of Done-line overreach as U-009/U-010
(development-plan.md claimed U-011 alone satisfies REQ-02's happy case,
but c_work_graph_plan's own contract explicitly excludes execution) and
flags a real, previously-invisible gap found while tracing REQ-02's actual
dependencies: no Unit in development-plan.md's 43-unit breakdown owns
c_artifact_write (03-C §3.7), despite REQ-02's happy AND failure cases both
requiring it directly. Not fixed here -- noted for whichever Unit (U-012/
U-013, most likely) needs it first.

15 new mocked stdlib unittest cases + 1 gated real test (re-run and passing
after the --json-schema fix). 98 total mocked tests across U-001-011, all
passing.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

trigger_org_node_full(): guards (TASK_NOT_READY -- wrong task.status or
required artifacts not validation_status=passed; CONCURRENCY_LIMIT_REACHED
-- only on the planner/queued transition, per 04-P's guard column;
USAGE_LIMIT_REACHED -- detected from the underlying claude error text,
best-effort against REQ-14's cited "hit your session limit" phrasing) then
calls U-005's trigger_org_node for the real subprocess call, then advances
task.status: queued->planning_brd_prd_sdd (planner), planning_brd_prd_sdd
->coding (coder, gated on brd/prd/sdd/benchmark all passed),
coding->validating (validator, gated on code_diff/test_report both passed).

Verified for real: all three Org Agents actually triggered in sequence
(chained claude --agent calls, ~15s/~$0.15 total) with task.status
advancing correctly at each step -- not just planner like U-004/U-005's
tests, all three personas confirmed invokable.

Also corrects two more development-plan.md issues found while scoping
this Unit, same class as U-002/U-009/U-036:
- U-010's Done line cited '架构文档 §11"手工调用验证"里程碑' -- that phrase
  doesn't exist in architecture doc §11 at all. The real source is 05
  §5.5's rollout step 2 ("手工调用验证...三个 Org Agent 分别可被触发"),
  whose "分别" (separately/respectively) is the load-bearing word: verify
  each CAN be triggered, not that a full artifact-validated chain runs
  end to end. That distinction is why this Unit's scope is exactly what
  got built -- not chasing planning_failed/coding_failed retries or
  validator PASS/FAIL routing, which need real Work Graph output (U-011+)
  or Org-level Loop routing (U-015/016, PH2) that don't exist yet.
- Added an explicit scope note spelling out exactly which 3 transitions
  this Unit owns vs. which parts of p_org_task_lifecycle it deliberately
  doesn't touch, so nobody has to re-derive this from the code later.

11 new mocked stdlib unittest cases + 1 gated real chained-trigger test.
87 total mocked tests across U-001-010, all passing; the real test
independently re-run and confirmed passing.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

create_task(): validates goal_text/project_id/task_type (INVALID_PARAM per
03-C §3.1), inserts the task row (status=queued, assigned_org_node=planner,
loop_round=0 -- 04-P p_org_task_lifecycle start_state) plus a
task_status_changed execution_event, both in one transaction (explicit
rollback on failure between the two inserts, tested by dropping
execution_event mid-call). c_project_create isn't reimplemented -- it's
already complete in U-002 (project_scaffold.create_project).

Corrects another Done-line overreach found while scoping this Unit:
development-plan.md claimed U-009 alone satisfies all three REQ-01 cases
(happy/boundary/failure), but the happy case ("依次触发 planner/coder/
validator...task.status 按状态机正确流转") needs c_org_node_trigger's full
implementation -- U-010's own Done criterion is almost that exact sentence,
and U-009 doesn't depend on U-010 or even touch org-node triggering. Same
class of issue as the U-002/U-036 corrections earlier this session (an
SDD/plan line claiming more than the Unit's actual dependencies support),
judged low-stakes enough (doesn't change what code to write, just when the
REQ-01 happy case gets marked satisfied) to fix directly rather than stop
and ask, per the graduated caution this session has been using. Boundary
(concurrency limit doesn't block task *creation*, only later scheduling)
and failure (INVALID_PARAM) are both fully real at U-009's scope and
tested; happy completes cumulatively once U-010 lands.

12 new stdlib unittest cases. 75 total across U-001-009, all passing.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

frontend/{index.html,app.js,style.css}: no build step, no npm, vanilla JS
hash router (05 §5.7). Real page content is PH3 (U-022+, starting from
docs/wireframes/); this is only the empty shell + working navigation the
Done criterion asks for.

backend_server.py serves the 3 files via an exact-path whitelist (not a
generic static-file/directory server) -- no path-traversal surface to
reason about for 3 fixed files.

Verified in an actual browser this session (navigated to localhost, clicked
all 3 nav links, screenshotted each) rather than only trusting stdlib
tests -- a Python-only test can't exercise real hashchange/DOM behavior
without a JS engine. Nav highlighting and content swap both confirmed
working, no page reload, no build step.

7 new stdlib unittest cases covering what the server is actually
responsible for (static files served with correct content-type, 404 on
unknown GET path). 66 total across U-001-008, all passing.

This closes out PH0 (U-001~U-008) -- all three Phase-acceptance criteria
now hold for real: target-project directory scaffold, 15-table schema with
constraint tests, and a real claude --agent round-trip.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Pure stdlib http.server (ThreadingHTTPServer), zero third-party deps --
enforced by an AST-based test that parses backend_server.py's own imports
and checks each against sys.stdlib_module_names, so this can't silently rot
if someone adds a dependency later (05 §5.7).

pick_folder(): macOS via `osascript 'choose folder'`; user-cancel maps to
{"selected_path": null} (03-C §3.-1: not an error), non-macOS or any other
osascript failure raises DialogUnavailableError -> HTTP 422
DIALOG_UNAVAILABLE. Windows/Linux native dialogs are PH3 (U-022).

Known gap, noted rather than papered over: the Done criterion asks for a
real popped dialog returning a real path, which needs an actual human
click -- not something this agent can do non-interactively. Verified
`osascript` itself is present and executes on this machine, and every code
path is covered by mocked tests, but the literal "did a human click a real
folder and get the right path back" check is still open for whoever runs
this interactively next.

12 new stdlib unittest cases (server start/stop, pick_folder success/
cancel/non-macOS/other-failure, the same 4 cases through the HTTP route,
zero-third-party-import check). 62 total across U-001-007 (60 real + 2
gated), all passing.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

write_workflow_script(): writes an orchestration script to
<target>/.claude/workflows/<name>.js -- versioned with the target project's
own git history, not graphlemon's (03-C §3.5 step 1).

execute_work_graph(): `claude -p 'Run /<name> with args {...}' --output-format
json --json-schema <schema> --permission-mode auto`, no `--bare` (A-1 spike:
skips credential discovery, breaks headless invocation), never
bypassPermissions. Returns structured_output. No budget-usage bookkeeping
or CLASSIFIER_BLOCKED handling yet -- U-012's job ("骨架" here).

Real-call coverage reuses spikes/a1-dynamic-workflows/workflow-under-test.js
verbatim (read from that file, not retyped) through the new wrapper --
"直接复用已验证的代码路径" per development-plan.md U-006, not just citing the
old spike as prior art. Confirmed passing for real: same result as the
original spike (sum=29 for [2,3,4]).

8 new mocked stdlib unittest cases + 1 gated real-agent test. 53 total
mocked tests across U-001-006, all passing; both gated real-agent tests
(U-004/U-005's planner call, U-006's reused spike workflow) re-verified
manually this session.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

trigger_org_node(org_node, prompt, target_folder_path): shells out
`claude --agent <org_node> -p ... --permission-mode auto` with cwd=target
project, parses the JSON result, raises OrgNodeTriggerError on nonzero exit
or is_error. Skeleton only, per development-plan.md ("不含完整状态机") --
TASK_NOT_READY/CONCURRENCY_LIMIT_REACHED/USAGE_LIMIT_REACHED handling is
U-010's job.

Also refactors test_agent_e2e.py (U-004) to call this real function instead
of its own ad-hoc subprocess invocation, so the gated real-agent test now
exercises U-005's actual production code path -- confirmed still passing
against a real `claude --agent planner` call.

5 new mocked stdlib unittest cases (CLAUDE.md Testing constraint: mock the
subprocess boundary in normal tests). 46 total across U-001-005, all
passing; the 1 gated real-agent test re-verified manually.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Resolves the A-2 assumption with a real `claude --agent planner -p` call
(--permission-mode auto, $0.05) against a deployed target project: the
migrated identity comes through correctly, not the default assistant
persona. Verdict is "部分成立" (partially confirmed), per the criterion
development-plan.md's U-004 line pre-registered for exactly this outcome --
Planner/Validator's SOUL.md migrated almost verbatim, but Coder's SOUL.md
needed real rewriting: the original hardcoded a live Feishu Base API token
(redacted, never committed), strongly bound its task-claiming to
`lark-cli`/Base status vocabulary, and told itself to shell out
`claude --dangerously-skip-permissions` (conflicts directly with 05
§5.3.1's `auto`-mode decision). All three replaced with graphlemon's own
task table / c_org_node_trigger / task.status vocabulary.

Also resolves a second, higher-stakes conflict found while scoping this
Unit: where does `.claude/agents/<org_node>.md` actually live? Architecture
doc §13.1 says graphlemon's own platform repo; 03-C §3.3 and 05 §5.1 (both
core SDD, not the architecture PRD) explicitly say the target project --
matching how Claude Code's --agent flag actually discovers subagents
(relative to invocation cwd). User confirmed 03-C/05 win. Resolved as two
different files, not a contradiction: agents/<org_node>/SOUL.md (+
validator/AGENTS.md) here is the canonical migrated source (01-M 1.9
agent_profile.soul_file_path, a singleton with no target-project
reference); src/agent_profile.py compiles that into the deployable
.claude/agents/<org_node>.md and writes it into a target project via
deploy_agent_definitions(). Does not reopen the U-002 correction from the
previous commit -- .graphlemon/ scaffolding and agent deployment are
different, non-conflicting concerns.

validator's tools/disallowedTools frontmatter encodes qa/AGENTS.md's
original workspace-permission contract (read-only on source, no git
mutation commands) as actual Claude Code permissions instead of prose.

14 new stdlib unittest cases (test_agent_profile.py, all real) + 1 gated
real-agent test (test_agent_e2e.py, RUN_REAL_AGENT_E2E=1, mirrors
graphloop's pattern per CLAUDE.md's Testing constraint -- this is the first
Unit that actually shells out to `claude`, so this is also where that
harness gets built, per the harness-ownership decision from the earlier
"什么还需要" review). 41 total tests across U-001-004, all passing.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>