Repository navigation
docs: correct stale layout text and complete the directory map - #559
Conversation
Removing infra and model left three sentences in module-layout.md that still state Infra rules in the present tense. They now say what changed. The architecture directory map gains the Apps delivery package, the Apps root compatibility exports and the modules added since the map was last updated, so every package on disk is listed. The contributor guide sent new models to harness/scenarios, whose one module runs any external command as a trial and is the same for every model; it now names the serving and Benchmark locations instead. The Harness, scenarios and views package docstrings now describe what the packages hold, including the search evidence that views writes for campaign import-search. No code changes. Agent-assisted-by: Claude Code / claude-opus-5-5 [effort=xhigh] (author) Signed-off-by: GUOGUO <xwdong1998@163.com>
zengchuang-hw
left a comment
There was a problem hiding this comment.
PASS
首轮独立检视:7a7ab84c3b5f26b20c098f439470d5bcfa972809 → 75b9872e78b70eb8602aedcf0e7c5146b5d0fe9f。本轮是 #498 后的文档收尾;#558 把真正改名单独跟踪,本 PR 没有更改路径、兼容边界或运行身份。
核对了目录图和当前源码:模型的服务与度量扩展点和模型无关的外部试验执行已分开,退役 Infra 的描述改为历史事实,Apps 视图的说明也明确记录了 campaign import-search 的写入。对三个 Python 文件分别解析 AST,去掉模块 docstring 后与 base 完全一致。未执行完整集成测试或查看 CI;未验证未来改名所需的配对 wheel 证据。
amy-why-3459
left a comment
There was a problem hiding this comment.
LGTM
Independent review of 7a7ab84c3b5f26b20c098f439470d5bcfa972809 → 75b9872e78b70eb8602aedcf0e7c5146b5d0fe9f (full diff). This is documentation follow-up to #498 only: no path, compatibility, or runtime-identity change.
I checked the directory map against the current main tree (7a7ab84c). The new Apps packages (delivery/, storage/, handoff/, workflows/, integrations/, legacy/), knowledge and harness additions, and the six view HTML templates are present. Spot-checked added map lines and the three package docstring updates against the modules themselves (apps/knowledge.py, knowledge/knowledge_find.py, harness/scenarios/external.py, and apps/views/search_attribution.import_search). The contributing.md / "Where to extend" table correctly sends new models to harness/serving and benchmark/metrics.py / validators, and describes external.py as model-independent.
Accuracy/performance comparison evidence is not applicable (docs and docstring-only changes). Architecture looks sound; the three __init__.py edits are docstring-only.
Merge should wait for green required checks. The harness/scenarios/ rename remains #558; if that rename lands, the new contributing.md and architecture table lines that still name harness/scenarios/external.py will need a follow-up.
There was a problem hiding this comment.
LGTM
PR review verdict — PASS
Reviewed base 7a7ab84c3b5f26b20c098f439470d5bcfa972809 and head 75b9872e78b70eb8602aedcf0e7c5146b5d0fe9f.
Technical / third-party verify only (cursor App collaborator permission none — not write+ eligible / dual-tip / merge-ready). Author is Dong1017 — this is independent third-party verify, not a Dong self-vote.
Checked
- Docs/layout sync after #498: architecture directory map adds missing modules; contributing/module-layout drop retired
infraguidance; harness/scenarios/apps.views package blurbs match trial-execution role. - Tip already has amy + zeng Approves.
Ready note
- Tip Ready
full-checkFAIL is the retired digest import intest_recovered_attempt_does_not_trust_a_rewritable_declaration— fixed on main by merged #556. Update branch onto that main and re-run Ready before steward merge.
…ry test #457 merged after #498 removed infra/, and its new recovery test still imported digest from inferrsi.infra.artifacts, so full-check fails on main. The value it forges is compared by harness.optimization, which takes digest from harness.artifact_primitives, so the test now imports that copy. Agent-assisted-by: Claude Code / claude-opus-5-5 [effort=xhigh] (author) Signed-off-by: GUOGUO <xwdong1998@163.com>
cce2159
|
Pushed The only red check on |
zengchuang-hw
left a comment
There was a problem hiding this comment.
NO PASS
增量检视:75b9872e78b70eb8602aedcf0e7c5146b5d0fe9f → cce2159c97d7a76d82620f05c1b19179c0e87eb6。上轮检视 的文档与 docstring 结论不变。本次唯一新增改动是 tests/test_project_explanation.py:497 的 digest 导入,不属于 #498 文档收尾范围。
批准前必做: 将文档改动同步到已包含 #556 的 main,保留主线的模块级 digest 导入,去掉本 PR 重复的函数内导入修正。我从 7a7ab84c、cce2159c 和 main@10ba7a55 读取了该文件,用隔离的 git merge-file 三方检查,确认该行发生冲突:主线删除了本地导入,PR 则替换该行。它不是文档问题,也不是 CI 状态问题。
未查看 CI、运行完整测试或实际合并。无其他非阻塞建议。
|
Reviewed the increment The only code change is I independently compared this file on the PR base Please rebase the documentation changes onto current main (already contains #556) and drop this in-function import fix, keeping main's module-level import. I am not approving this head. Accuracy/performance comparison evidence is not applicable (docs plus a test-import change). The earlier documentation conclusions on |
#556 fixed the same stale import in tests/test_project_explanation.py with a module-level import from harness.artifact_primitives. The merge keeps main's file, so this branch no longer changes that test and its diff is the docs and docstrings only. Agent-assisted-by: Claude Code / claude-opus-5-5 [effort=xhigh] (author) Signed-off-by: GUOGUO <xwdong1998@163.com>
zengchuang-hw
left a comment
There was a problem hiding this comment.
PASS
增量检视 cce2159c97d7a76d82620f05c1b19179c0e87eb6 → caf2517c65f662b16e680023d028e14881357cb3。上轮 NO PASS 的重复导入和冲突已解决:新提交确实包含 main@10ba7a55,tests/test_project_explanation.py 与主线逐字节一致,保留了 #556 的模块级导入。相对主线只剩原先已通过的六个文档或 docstring 文件,均与 75b9872 逐字节一致。
我重新比对了精确提交、文件内容和三个 Python 文件去除模块 docstring 后的 AST,确认没有运行逻辑变化。没有新的阻塞项或非阻塞建议。未查看 CI 或运行完整测试;#558 的实际改名不在本 PR 范围内。
|
Merged |
|
@amy-why-3459 PTAL |
amy-why-3459
left a comment
There was a problem hiding this comment.
LGTM
PR review verdict — PASS
Reviewed the increment cce2159c97d7a76d82620f05c1b19179c0e87eb6 → caf2517c65f662b16e680023d028e14881357cb3 (base 10ba7a5500bcd95949219b138f9ada5033b922b0).
The previous blocker is resolved: this head contains main@10ba7a55 / #556, tests/test_project_explanation.py is byte-identical to that main (and to current main@3bf0ae9f), and the file is no longer in the PR diff. The remaining six documentation/docstring files are byte-identical to the previously reviewed 75b9872e docs-only revision. GitHub reports mergeable: true against current main. Ready-lane checks on this head have completed successfully.
Accuracy/performance comparison evidence is not applicable (docs and docstrings only). I did not rerun the test suite locally. I also saw the PTAL in 7063057382; this approval is bound to caf2517c.
Main gained #559 as a squash commit and #541's release workflow. The docs and the scenarios initializer conflicted only because this branch already carried #559's text with the rename on top; the merge keeps this branch's versions, and trials/__init__.py equals main's scenarios/__init__.py byte for byte. release.yml merged cleanly and still passes INFERRSI_PRE_CHANGE_SESSIONS to its one check_installed run. Agent-assisted-by: Claude Code / claude-opus-5-5 [effort=xhigh] (author) Signed-off-by: GUOGUO <xwdong1998@163.com>
Governing Issue and delivery slice
Refs #498
Refs #558
infra/andmodel/packages into their owning modules #498, approved by the goal owner (Dong1017) on 2026-11-21 as step 1 of the directory clean-up plan. task: renameharness/scenarios/to match its trial-execution role #558 tracks theharness/scenarios/rename separately.docs/module-layout.md: three sentences that still stated Infra rules in the present tense after task: retire theinfra/andmodel/packages into their owning modules #498 removedinfra/now say what changed (R2 paragraph, the closure rule, the second-importer list).docs/architecture.mddirectory map: addsapps/delivery/,apps/knowledge.py, the Apps root compatibility exports (archive.py…skill_evaluation.py,cannbot/,mindie/),integrations/cannbot/round_knowledge.py, the views, Knowledge, Harness and Benchmark modules added since the map was last updated, and the full set of view templates. Each new line is taken from that module's own docstring.docs/contributing.mdand the "Where to extend" table: new models go toharness/servingandbenchmark/metrics.pywithbenchmark/validators.harness/scenarios/external.pyruns any external command as one trial and is the same for every model (external.py:1).harness/__init__.py(it re-exports verdict labels from contracts and does no reporting itself),harness/scenarios/__init__.py, andapps/views/__init__.py, which said the package writes nothing althoughsearch_attribution.import_searchwrites the search evidence forcampaign import-search.inferrsi.infra.artifactsimport intests/test_project_explanation.pyis fixed onmainby Use the Harness digest in the interrupted recovery test #556; this branch keepsmain's file. Theharness/scenarios/rename is task: renameharness/scenarios/to match its trial-execution role #558. Theapps/views/split is a suggestion on task: Apps 应用层详细设计与迁移计划(CLI / Python / Plugin) #371 for the Apps owner.Knowledge authorship declaration
No path protected by
docs/knowledge-model-policy.mdchanges. The declaration below is disclosure only.Changed artifacts and evidence
docs/module-layout.mdRETIRED_MODULESintests/_gate_policy_support.py:103is the two package namesdocs/architecture.mdmain7a7ab84; descriptions come from module docstringsdocs/contributing.mdharness/scenarios/external.py:1: "independent of model and device"__init__.pydocstrings_runtime_identity()or the skill-eval controller hashesChecks and dependencies
git diff --checkis clean. Locally on Windows,tests/test_architecture_closed_loop.py,tests/test_target_module_import_gate.py,tests/test_package_root_inventory.pyandtests/test_template_layout.py, filtered to layout, doc, root and template tests: 46 passed. The tests that readmodule-layout.mdlook for phrases this PR keeps.test_template_layout.py::test_maintained_documentation_links_resolvefails locally because this checkout is sparse and has noevidence/tree; it fails the same way on unmodifiedmainhere. CI has the full checkout. The full suite was not run locally; CIfull-checkcovers it.Remaining work and Issue closure
harness/scenarios/to match its trial-execution role #558 (rename). task: retire theinfra/andmodel/packages into their owning modules #498 is already closed.🤖 Generated with Claude Code