@elvis1513/auto-coding-skill 4.2.7 → 4.3.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -9,6 +9,14 @@ The Skill is a selectable guardrail, not a command sequence that must run for
9
9
  every task. The model skips machinery whose expected benefit does not exceed its
10
10
  cost; read-only work and obvious small clean-checkout changes normally stay direct.
11
11
 
12
+ Version 4.3.0 separates exact managed defaults from a byte-stable project
13
+ configuration overlay, migrates legacy specialization with semantic proof, and
14
+ adds project-owned shared-Skill feedback inboxes with bounded read-only
15
+ multi-project triage. Upgrades now use recoverable, owner-bound transactions.
16
+ Version 4.2.8 makes the supervised Reviewer observable and recoverable without
17
+ weakening review semantics: it streams private event metadata, distinguishes
18
+ startup failure from analysis timeout, retries one no-event startup, and supports
19
+ an audited fingerprint-bound user override that is never reported as approval.
12
20
  Version 4.2.7 freezes every pre-commit Reviewer diff as a private, SHA-256-bound
13
21
  Git-local patch, including staged, unstaged, untracked, deleted, mode, symlink,
14
22
  and binary changes. Version 4.2.6 makes the managed structure/optimization description defer to
@@ -59,20 +67,70 @@ The project install contains:
59
67
  .agents/skills/auto-coding-skill/
60
68
  .agents/agents/
61
69
  docs/ENGINEERING.md
70
+ docs/project/auto-coding-skill.yaml
62
71
  docs/tools/autopipeline/ap.py
63
72
  AGENTS.md (fully managed canonical repository contract)
64
73
  ```
65
74
 
66
75
  `autocoding init` also installs the exact shared documentation tree under
67
- `docs/{architecture,bugs,deployment,design,interfaces,project,reviews,testing}`.
76
+ `docs/{architecture,bugs,deployment,design,interfaces,project,reviews,skill-feedback,testing}`.
68
77
  Managed templates are replaced; the scaffolded project structure standard,
69
78
  mutable project docs, and valid architecture, ADR, interface, DD, review, or
70
- deploy-record documents are preserved; unrelated
79
+ deploy-record documents plus `docs/skill-feedback/reports/*.md` are preserved;
80
+ unrelated
71
81
  directories are archived under `.agents/archive/`. Re-running init is the complete upgrade operation;
72
- `sync` and project-local `upgrade` are compatibility commands, not required steps.
82
+ `sync` is the explicit multi-project equivalent. Project-local
83
+ `ap.py upgrade --dry-run` remains a read-only legacy diagnostic; its old write
84
+ mode is retired so upgrades cannot bypass the transactional installer.
85
+ Multi-project sync completes all read-only preflights before its first write;
86
+ after that boundary, each project has its own recoverable transaction, so a
87
+ later host I/O failure does not roll back projects that already completed.
88
+
89
+ `docs/ENGINEERING.md` is the exact managed default layer. Project-specific
90
+ access, commands, validation routes, risk rules, structure policy, and other
91
+ supported values belong in `docs/project/auto-coding-skill.yaml`. Runtime commands
92
+ recursively overlay project mappings on managed defaults; project scalar and list
93
+ values replace defaults, while code-enforced safety invariants remain mandatory.
94
+ Ordinary init, sync, and upgrade operations never rewrite an existing overlay.
95
+ The first compatible upgrade extracts the semantic difference from the previous
96
+ managed default, verifies equivalence, and creates the overlay before replacing
97
+ managed files.
98
+
99
+ ## Shared Skill feedback
100
+
101
+ Every initialized project receives:
73
102
 
74
- Fill the project/Jenkins/GitLab/Nexus URL, username, and password fields under
75
- `access.*`, then configure one real fast validation command and run:
103
+ ```text
104
+ docs/skill-feedback/
105
+ ├── README.md
106
+ ├── _TEMPLATE-SKILL-FEEDBACK.md
107
+ └── reports/ # created when the first report is added
108
+ └── YYYY-MM-DD-<slug>-<8-hex>.md
109
+ ```
110
+
111
+ The README and template are managed; reports are project-owned and upgrades
112
+ preserve them byte-for-byte. A report is a candidate observation only. Create one
113
+ after distinguishing managed Skill behavior from project `risk.rules`, validation
114
+ routes, access values, structure policy, business code, and environment failures.
115
+ Do not automatically turn test failures or Reviewer findings into reports, and do
116
+ not include credentials, customer/device data, complete patches, Reviewer
117
+ artifacts, raw logs, or absolute user paths.
118
+
119
+ Periodic triage is explicit and read-only:
120
+
121
+ ```bash
122
+ autocoding feedback --projects /path/geestock,/path/geesight,/path/xjmate --json
123
+ ```
124
+
125
+ The collector scans only bounded report metadata from the listed projects,
126
+ executes nothing from the documents, changes no repository, and groups exact
127
+ root-cause signatures. Human triage still decides whether an observation is a
128
+ shared defect, project configuration, environment issue, duplicate, or missing
129
+ evidence before fixes and a release are planned.
130
+
131
+ Fill the project/Jenkins/GitLab/Nexus URL, username, and password overrides under
132
+ `access.*` in `docs/project/auto-coding-skill.yaml`, then configure one real fast
133
+ validation command and run:
76
134
 
77
135
  ```bash
78
136
  python3 docs/tools/autopipeline/ap.py doctor
@@ -236,16 +294,15 @@ never triggers a broader fallback.
236
294
  - Parallel fixers require separate worktrees and non-overlapping paths.
237
295
  - Reviewer fingerprint approval is required for high-risk, cross-module,
238
296
  explicitly configured, or parallel integration work.
239
- - The `reviewer` Agent remains `xhigh`. A focused review gets one 90-second pass
240
- over the supplied stable diff and relevant evidence. Timeout produces
241
- `blocked`; do not call `task-review` for that result.
297
+ - Focused review uses `high` effort and a 150-second hard wall over the supplied
298
+ stable diff and evidence. Deep review uses `xhigh` and 360 seconds.
242
299
  - Cross-module or parallel work and API, auth, database, payment, file-transfer,
243
- gateway, or production-configuration boundaries use 300-second deep review.
300
+ gateway, or production-configuration boundaries use deep review.
244
301
  Fixing Reviewer JSON formatting reuses the same analysis and never starts
245
302
  another substantive review.
246
303
  - `review-assignment` writes the validated assignment under Git-local state with
247
304
  the exact fingerprint, HEAD, scope revision, owning fixer, review depth, and
248
- 90/300-second deadline. Before that deadline begins, it freezes the exact
305
+ 150/360-second deadline. Before that deadline begins, it freezes the exact
249
306
  task-owned working-tree state into a mode-0600 binary patch and binds its
250
307
  canonical path, format, and SHA-256 into the assignment. Reissuing the same
251
308
  live assignment is idempotent. Snapshot inputs and emitted patches above
@@ -255,11 +312,20 @@ never triggers a broader fallback.
255
312
  rejected by `task-review`.
256
313
  - `review-run` is the default executable path: it creates/reuses that assignment,
257
314
  starts a separate `codex exec` session in read-only mode, removes the parent
258
- lifecycle identity, and supervises the full process group against the original
259
- remaining deadline. Timeout performs TERM then KILL, records a Git-local blocked
260
- receipt, and cannot be converted into approval. Plain `review-assignment` remains
315
+ lifecycle identity, consumes JSON events incrementally, and supervises the full
316
+ process group against the original remaining deadline. If no semantic event
317
+ arrives within 30 seconds it retries the same immutable assignment once. After
318
+ analysis starts it never retries. Private schema-2 receipts record only bounded,
319
+ allowlisted event/diagnostic categories, byte counts, and hashes. Plain `review-assignment` remains
261
320
  available only when another host can enforce the deadline itself; it does not
262
- stop an in-app subagent.
321
+ stop an in-app subagent. A descendant that deliberately creates a new POSIX
322
+ session is outside portable process-group termination, but cannot keep the
323
+ non-blocking event readers or Reviewer deadline open.
324
+ - Exhausted `runtime-unavailable` or `analysis-timed-out` states remain blocked by
325
+ default. `review-runtime-override` can record explicit user authorization bound
326
+ to the exact fingerprint, assignment, artifact, and runtime receipt. Its verdict
327
+ is `runtime-bypassed`, never `approved`; real blocked/changes-requested results
328
+ cannot use this path.
263
329
  - DD/ADR is created only for lasting cross-module, API, data, security,
264
330
  deployment, or key user-flow decisions.
265
331
  - Historical debt does not block product work unless the current change worsens it.
@@ -308,6 +374,47 @@ model overrides, complete project `risk.rules`, supported project/access/
308
374
  concurrency/route/structure values, and an existing project-owned structure
309
375
  standard byte-for-byte. Removed content is archived outside active docs.
310
376
 
377
+ ## What changed in 4.3.0
378
+
379
+ - Made `docs/ENGINEERING.md` the exact managed default layer and added
380
+ `docs/project/auto-coding-skill.yaml` as the higher-priority project-owned
381
+ overlay. Mappings merge recursively; explicit project scalars and lists replace
382
+ defaults, while managed workflow identity and code-enforced safety invariants
383
+ remain protected.
384
+ - Added one-time semantic migration from the previous manifest-bound default.
385
+ Initialization creates the overlay without overwriting an existing one, proves
386
+ the reconstructed effective configuration is equivalent, archives the legacy
387
+ document, and then converges the managed default byte-for-byte.
388
+ - Added `docs/skill-feedback/` with managed guidance/template and preserved
389
+ project-owned reports. `autocoding feedback --projects ... --json` performs an
390
+ explicit bounded metadata-only read and groups exact root-cause signatures for
391
+ later human triage.
392
+ - Replaced partial upgrade writes with owner-bound, recoverable installation
393
+ transactions, staged/installed hash checks, active-installer rejection, safe
394
+ project-file operations, and fail-closed effective-configuration validation.
395
+ Project-local `ap.py upgrade --write` is retired in favor of transactional
396
+ `autocoding init` or explicit multi-project `sync`.
397
+ - Batched protected executable-mode convergence so the safer installer retains
398
+ practical upgrade latency while preserving symlink/reparse checks.
399
+
400
+ ## What changed in 4.2.8
401
+
402
+ - Replaced end-only Reviewer output collection with bounded JSONL event streaming.
403
+ Schema-2 Git-local receipts now record CLI/model/effort identity, attempt phases,
404
+ first/last event times, event-log hashes, diagnostic categories, and byte counts
405
+ without persisting prompt, patch, stderr, tool output, or model content.
406
+ - Split runtime startup failure from analysis timeout. A no-semantic-event startup
407
+ is capped at 30 seconds and retried once with the same assignment, fingerprint,
408
+ artifact, command, and absolute deadline; an analysis that started is never
409
+ repeated. Focused review now uses `high/150s`, while deep uses `xhigh/360s`.
410
+ - Added `review-runtime-override` for explicit user-authorized delivery after an
411
+ exhausted runtime-unavailable or analysis-timeout failure. The private override
412
+ binds all immutable review evidence and records `runtime-bypassed`, never
413
+ `approved`; failure-time hashes reject pre-authorization tampering, and a
414
+ substantive Reviewer result observed before timeout remains non-bypassable.
415
+ - Restricted the hidden custom runner to explicit test harnesses and closed spawn
416
+ failure, unbounded-output, silent-startup, and lost-diagnostics paths.
417
+
311
418
  ## What changed in 4.2.7
312
419
 
313
420
  - Replaced live pre-commit diff discovery with an immutable Git-local patch
@@ -8,8 +8,8 @@ developer_instructions = """
8
8
 
9
9
  工作要求:
10
10
  1. 只审查主代理给定的 task_id、base_sha、diff_fingerprint、diff_artifact_path、diff_artifact_sha256 和 owning_fixer;开始前必须运行 `python3 docs/tools/autopipeline/ap.py review-artifact --file <assignment.json>`,以验证 0600 权限和 SHA-256 并读取冻结补丁。不得用实时 `git diff` 或 `diff_base..diff_head` 替代该 artifact。
11
- 2. 若主代理指定 focused review,只对给定的稳定 diff 和相关 evidence 做一次实质分析,不扩展为全库扫描。运行器会在 90 秒截止点终止整个 Reviewer 进程组;若你在终止前已发现无法继续,则返回 `status=blocked`、`verdict=blocked`。
12
- 3. 跨模块、并行 writer,以及 API、鉴权、数据库、支付、文件传输、网关、生产配置边界使用 deep review,预算 300 秒;仍然只审查主代理交付的范围。
11
+ 2. 若主代理指定 focused review,只对给定的稳定 diff 和相关 evidence 做一次实质分析,不扩展为全库扫描。focused 运行使用 high effort 和 150 秒硬截止;若你在终止前已发现无法继续,则返回 `status=blocked`、`verdict=blocked`。
12
+ 3. 跨模块、并行 writer,以及 API、鉴权、数据库、支付、文件传输、网关、生产配置边界使用 deep review,采用 xhigh effort 和 360 秒硬截止;仍然只审查主代理交付的范围。运行器仅可在 30 秒内没有语义事件时复用同一 assignment 自动重启一次;一旦分析已开始不得重试实质审查。
13
13
  4. 优先检查:正确性、安全性、边界条件、异常路径、并发风险、事务一致性、缓存一致性、兼容性。
14
14
  5. 结构/优化 review 仅在已有 baseline/backlog 时读取;可选文档缺失不构成 blocker,不重复报告 accepted debt。
15
15
  6. 检查已验证的冻结 diff artifact 和已执行验证是否一致;该 artifact 统一包含 staged、unstaged、untracked、删除、模式变化和二进制变更。普通任务不要求 taskbook、closure 或 evidence 文档。
@@ -1 +1 @@
1
- {"schema_version":1,"skill_version":"4.2.7","manifest_path":".agents/managed-install.json","entries":[{"path":".agents/agents/browser-debugger.toml","source":"agents/browser-debugger.toml","ownership":"normalized-agent","sha256":"40727952944917eb0b573bb96dca2906c27bc794737f3b6b392b56df7b4208aa","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/agents/docs-researcher.toml","source":"agents/docs-researcher.toml","ownership":"normalized-agent","sha256":"aa1fed6a6a1cde282fc152c7a88b8e097a6c53bcf447d35978607d594e3cb3e4","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/agents/explorer.toml","source":"agents/explorer.toml","ownership":"normalized-agent","sha256":"4cd89888acf7c0c04fbbcd0b9a0f346c6efed42fb723ec2037af9331a84a1cac","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/agents/fixer.toml","source":"agents/fixer.toml","ownership":"normalized-agent","sha256":"7bc00a45b9d1bcc90b8b0f06d773556eba1d52effff664472c3cb81073f6abfb","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/agents/reviewer.toml","source":"agents/reviewer.toml","ownership":"normalized-agent","sha256":"0e5939138dd88d8ef13267bae68f0547d449602ff7b51879a4fc24ba64f64edc","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/SKILL.md","source":"skill/SKILL.md","ownership":"exact","sha256":"e492798ae0113305e3a27139ab431e7b0bd62323733797440d6615696ce74e75","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/data/contracts/orchestration-v1.schema.json","source":"skill/data/contracts/orchestration-v1.schema.json","ownership":"exact","sha256":"65a8c0d222b26f224794ec106101f903f0e3a357b249c2df22c5908a476a6835","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/data/policies/workflow-migrations-v1.json","source":"skill/data/policies/workflow-migrations-v1.json","ownership":"exact","sha256":"e47fd99c449383394078ae9c1d951aa25e9a6c6f50b626489c395ff0729c33de","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/data/templates/ENGINEERING.md","source":"skill/data/templates/ENGINEERING.md","ownership":"exact","sha256":"270131e1ecd53ccfa5325d240aac9b13529a3fb6f23b985626c3f9a10852a1ab","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/data/templates/bridges/AGENTS.md","source":"skill/data/templates/bridges/AGENTS.md","ownership":"exact","sha256":"f825db96bf19f1d08ed185a04df60745bc425f86dc2a3f59f362a6cdfe983e06","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/data/templates/tools/ap.py","source":"skill/data/templates/tools/ap.py","ownership":"exact","sha256":"c352fbf2cbcb9692a34fd952d8541b1cbe6d475c6e11e64500a1f6b5283a00be","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/requirements.txt","source":"skill/requirements.txt","ownership":"exact","sha256":"0969da99a0bc2a1b71ed50584560f4588a37567ac63af3ddbaf3c4617ca5621a","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/scripts/ap.py","source":"skill/scripts/ap.py","ownership":"exact","sha256":"57fb42399cf4e1f7ca8cc2e1142edb903a1b891b31a796e9d4aba51c86a0a016","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/scripts/core.py","source":"skill/scripts/core.py","ownership":"exact","sha256":"640c37e8b274d90d6b7765d4243e054360d5a5a85b3150fbfe6449527c1cd28f","executable":true,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/scripts/http_checks.py","source":"skill/scripts/http_checks.py","ownership":"exact","sha256":"5620e187a0129f135621425a84b3a4e8f31137263c7e549a36a801fc36ff54ea","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/scripts/install_integrity.py","source":"skill/scripts/install_integrity.py","ownership":"exact","sha256":"a9ca7bd67893bbd2d7b35ece30a40457f0cbcb9ae01580cfc5daca2cd96fd676","executable":false,"scope":"shared","version":"4.2.7"},{"path":".agents/skills/auto-coding-skill/scripts/scaffold_templates.py","source":"skill/scripts/scaffold_templates.py","ownership":"exact","sha256":"6f0ee92eff2d5db9627a64b332738c9c917a93cc559d5bbacf771c5966215821","executable":false,"scope":"shared","version":"4.2.7"},{"path":"AGENTS.md","source":"skill/data/templates/bridges/AGENTS.md","ownership":"exact","sha256":"f825db96bf19f1d08ed185a04df60745bc425f86dc2a3f59f362a6cdfe983e06","executable":false,"scope":"project","version":"4.2.7"},{"path":"docs/ENGINEERING.md","source":"skill/data/templates/ENGINEERING.md#managed-workflow","ownership":"managed-workflow","sha256":"fdd3ac3689a0436bc439f01f76a8e8d627450cebdb700d2107badbbfec02b7b6","executable":false,"scope":"project","version":"4.2.7"},{"path":"docs/architecture/adr/_TEMPLATE-ADR.md","source":"generated:scaffold_templates:docs/architecture/adr/_TEMPLATE-ADR.md","ownership":"exact","sha256":"88c89065ae0f8555412dbe096c05342152efc7becae8d8981e6442cdfee8e672","executable":false,"scope":"project","version":"4.2.7"},{"path":"docs/deployment/deploy-records/_TEMPLATE-DEPLOY-RECORD.md","source":"generated:scaffold_templates:docs/deployment/deploy-records/_TEMPLATE-DEPLOY-RECORD.md","ownership":"exact","sha256":"f2c17c0febc322b0ad50fb9d6f7e64e6c0cfb15c8a3e445a99da40f0a0dd1901","executable":false,"scope":"project","version":"4.2.7"},{"path":"docs/design/_TEMPLATE-DD.md","source":"generated:scaffold_templates:docs/design/_TEMPLATE-DD.md","ownership":"exact","sha256":"6ca022d127bfff68521f466b60bace4f1206baf0913daa855b9f751e74198db3","executable":false,"scope":"project","version":"4.2.7"},{"path":"docs/reviews/_TEMPLATE-REVIEW.md","source":"generated:scaffold_templates:docs/reviews/_TEMPLATE-REVIEW.md","ownership":"exact","sha256":"134f0d1bd3c7a317ef3cf4c2aea9f0539f3d53dccd5bcdabf6915b3629ef8802","executable":false,"scope":"project","version":"4.2.7"},{"path":"docs/tools/autopipeline/ap.py","source":"skill/data/templates/tools/ap.py","ownership":"exact","sha256":"c352fbf2cbcb9692a34fd952d8541b1cbe6d475c6e11e64500a1f6b5283a00be","executable":false,"scope":"project","version":"4.2.7"}],"managed_namespaces":[{"path":".agents/skills/auto-coding-skill","scope":"shared"},{"path":"docs/tools/autopipeline","scope":"project"}],"preserved":[".agents/archive/**",".agents/agents/* (except paths listed in entries)","docs/project/**","docs/architecture/structure-standard.md","docs/architecture/adr/[0-9]*.md","docs/design/*.md","docs/interfaces/*.md","docs/deployment/deploy-records/*.md","docs/reviews/*.md"]}
1
+ {"schema_version":1,"skill_version":"4.3.0","manifest_path":".agents/managed-install.json","entries":[{"path":".agents/agents/browser-debugger.toml","source":"agents/browser-debugger.toml","ownership":"normalized-agent","sha256":"40727952944917eb0b573bb96dca2906c27bc794737f3b6b392b56df7b4208aa","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/agents/docs-researcher.toml","source":"agents/docs-researcher.toml","ownership":"normalized-agent","sha256":"aa1fed6a6a1cde282fc152c7a88b8e097a6c53bcf447d35978607d594e3cb3e4","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/agents/explorer.toml","source":"agents/explorer.toml","ownership":"normalized-agent","sha256":"4cd89888acf7c0c04fbbcd0b9a0f346c6efed42fb723ec2037af9331a84a1cac","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/agents/fixer.toml","source":"agents/fixer.toml","ownership":"normalized-agent","sha256":"7bc00a45b9d1bcc90b8b0f06d773556eba1d52effff664472c3cb81073f6abfb","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/agents/reviewer.toml","source":"agents/reviewer.toml","ownership":"normalized-agent","sha256":"e1ad676e608ffee04d782571fb9bfbc787f69c100941b1811d955ef80439db36","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/SKILL.md","source":"skill/SKILL.md","ownership":"exact","sha256":"d142951aa5cd7062ec467421654c7cda0a9dac2262eedc97a92590428b3a21e7","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/data/contracts/orchestration-v1.schema.json","source":"skill/data/contracts/orchestration-v1.schema.json","ownership":"exact","sha256":"65a8c0d222b26f224794ec106101f903f0e3a357b249c2df22c5908a476a6835","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/data/policies/workflow-migrations-v1.json","source":"skill/data/policies/workflow-migrations-v1.json","ownership":"exact","sha256":"b3d1018ad1778045cd6b329a2591576f8ab4e3ec4ff9b01036ffaeb327ecc8db","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/data/templates/ENGINEERING.md","source":"skill/data/templates/ENGINEERING.md","ownership":"exact","sha256":"c31b61084a4a20d95f2b5a6168efce596e9487d4b47d1f979cc4da915b13928d","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/data/templates/bridges/AGENTS.md","source":"skill/data/templates/bridges/AGENTS.md","ownership":"exact","sha256":"2c21fd3ecfd34334aeadf203247150e2ec0abfa7f11ec0cb4f94bfe95d6a4d73","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/data/templates/tools/ap.py","source":"skill/data/templates/tools/ap.py","ownership":"exact","sha256":"c352fbf2cbcb9692a34fd952d8541b1cbe6d475c6e11e64500a1f6b5283a00be","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/requirements.txt","source":"skill/requirements.txt","ownership":"exact","sha256":"0969da99a0bc2a1b71ed50584560f4588a37567ac63af3ddbaf3c4617ca5621a","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/scripts/ap.py","source":"skill/scripts/ap.py","ownership":"exact","sha256":"e893422f616017c825b2ec3e109ee5fbe96c6262c9ada31f9e92663d23408008","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/scripts/core.py","source":"skill/scripts/core.py","ownership":"exact","sha256":"caefca925929bad840c3471b86332d4be48b5e19ad0ff79fde1dcedef0297c16","executable":true,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/scripts/http_checks.py","source":"skill/scripts/http_checks.py","ownership":"exact","sha256":"5620e187a0129f135621425a84b3a4e8f31137263c7e549a36a801fc36ff54ea","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/scripts/install_integrity.py","source":"skill/scripts/install_integrity.py","ownership":"exact","sha256":"44eb23b2bfbfc854be9c23e5ff7a4f133a32365316af976844ed129a3a69d2e3","executable":false,"scope":"shared","version":"4.3.0"},{"path":".agents/skills/auto-coding-skill/scripts/scaffold_templates.py","source":"skill/scripts/scaffold_templates.py","ownership":"exact","sha256":"ff1fe25c65da0f93baf84f358d1117ae038f16e996a443b73cc06d11699aa36d","executable":false,"scope":"shared","version":"4.3.0"},{"path":"AGENTS.md","source":"skill/data/templates/bridges/AGENTS.md","ownership":"exact","sha256":"2c21fd3ecfd34334aeadf203247150e2ec0abfa7f11ec0cb4f94bfe95d6a4d73","executable":false,"scope":"project","version":"4.3.0"},{"path":"docs/ENGINEERING.md","source":"skill/data/templates/ENGINEERING.md","ownership":"exact","sha256":"c31b61084a4a20d95f2b5a6168efce596e9487d4b47d1f979cc4da915b13928d","executable":false,"scope":"project","version":"4.3.0"},{"path":"docs/architecture/adr/_TEMPLATE-ADR.md","source":"generated:scaffold_templates:docs/architecture/adr/_TEMPLATE-ADR.md","ownership":"exact","sha256":"88c89065ae0f8555412dbe096c05342152efc7becae8d8981e6442cdfee8e672","executable":false,"scope":"project","version":"4.3.0"},{"path":"docs/deployment/deploy-records/_TEMPLATE-DEPLOY-RECORD.md","source":"generated:scaffold_templates:docs/deployment/deploy-records/_TEMPLATE-DEPLOY-RECORD.md","ownership":"exact","sha256":"f2c17c0febc322b0ad50fb9d6f7e64e6c0cfb15c8a3e445a99da40f0a0dd1901","executable":false,"scope":"project","version":"4.3.0"},{"path":"docs/design/_TEMPLATE-DD.md","source":"generated:scaffold_templates:docs/design/_TEMPLATE-DD.md","ownership":"exact","sha256":"6ca022d127bfff68521f466b60bace4f1206baf0913daa855b9f751e74198db3","executable":false,"scope":"project","version":"4.3.0"},{"path":"docs/reviews/_TEMPLATE-REVIEW.md","source":"generated:scaffold_templates:docs/reviews/_TEMPLATE-REVIEW.md","ownership":"exact","sha256":"134f0d1bd3c7a317ef3cf4c2aea9f0539f3d53dccd5bcdabf6915b3629ef8802","executable":false,"scope":"project","version":"4.3.0"},{"path":"docs/skill-feedback/README.md","source":"generated:scaffold_templates:docs/skill-feedback/README.md","ownership":"exact","sha256":"7beea4818091cef49fd41022e4f423adf1c917525d694d137eb807ba69d2f920","executable":false,"scope":"project","version":"4.3.0"},{"path":"docs/skill-feedback/_TEMPLATE-SKILL-FEEDBACK.md","source":"generated:scaffold_templates:docs/skill-feedback/_TEMPLATE-SKILL-FEEDBACK.md","ownership":"exact","sha256":"d72a65d922c8b0609050e68d509b983eb18841c71b53dd5db4d9b75b36750785","executable":false,"scope":"project","version":"4.3.0"},{"path":"docs/tools/autopipeline/ap.py","source":"skill/data/templates/tools/ap.py","ownership":"exact","sha256":"c352fbf2cbcb9692a34fd952d8541b1cbe6d475c6e11e64500a1f6b5283a00be","executable":false,"scope":"project","version":"4.3.0"}],"managed_namespaces":[{"path":".agents/skills/auto-coding-skill","scope":"shared"},{"path":"docs/tools/autopipeline","scope":"project"}],"preserved":[".agents/archive/**",".agents/agents/* (except paths listed in entries)","docs/project/**","docs/project/auto-coding-skill.yaml","docs/architecture/structure-standard.md","docs/architecture/adr/[0-9]*.md","docs/design/*.md","docs/interfaces/*.md","docs/deployment/deploy-records/*.md","docs/reviews/*.md","docs/skill-feedback/reports/*.md"]}
@@ -7,9 +7,9 @@ description: Optional delivery-first engineering workflow for repository changes
7
7
 
8
8
  ## Use the two authorities
9
9
 
10
- Read root `AGENTS.md` for the shared behavioral protocol,
11
- `docs/ENGINEERING.md` for access/risk/validation configuration, and
12
- `docs/project/` for durable project facts.
10
+ Read root `AGENTS.md` for the shared behavioral protocol, the effective
11
+ configuration formed from managed `docs/ENGINEERING.md` defaults plus
12
+ `docs/project/auto-coding-skill.yaml`, and `docs/project/` for durable facts.
13
13
  Do not reconstruct or duplicate those rules from historical task documents.
14
14
 
15
15
  Normal delivery is:
@@ -52,13 +52,12 @@ Each parallel writer uses a task ID, worktree, lease, dependency SHAs, and disti
52
52
  with `agent-contract-check`; parallel fixers require reclassification with
53
53
  `--writers >1`.
54
54
 
55
- Use `reviewer` at `xhigh`: focused gets 90 seconds; parallel/cross-module or sensitive boundaries get 300. `review-run` creates the assignment, launches a separate read-only
56
- Codex process without the lifecycle-owner identity, terminates its process group at the deadline, and records only an exact HEAD/scope/fingerprint match. Before the deadline,
57
- `review-assignment` freezes staged, unstaged, untracked, deleted, mode, symlink, and binary changes into a mode-0600 Git-local patch with assignment-bound path, format, and SHA-256.
58
- The Reviewer must run `python3 docs/tools/autopipeline/ap.py review-artifact --file <assignment.json>` and review that emitted patch instead of reconstructing a live Git diff. Timeout is `blocked`.
59
- `review-assignment` alone requires another deadline-capable host and cannot stop an
60
- in-app subagent. `agent-result-template` supplies all 16 fields; `review-run` safely
61
- normalizes presentation fields from the same analysis.
55
+ Focused review uses `high`/150 seconds; deep review uses `xhigh`/360 seconds. `review-run` launches a separate read-only Codex process, streams allowlisted event metadata to a private Git-local log, stops the process group at deadline, and binds the result to the exact HEAD/scope/fingerprint.
56
+ No semantic event within 30 seconds retries the same assignment once; started analysis is never retried. `review-assignment` freezes every working-tree change into a mode-0600, assignment-bound, SHA-256 patch.
57
+ The Reviewer reads it with `review-artifact --file <assignment.json>`, never a reconstructed live diff. `review-assignment` alone needs another deadline-capable host; `agent-result-template` supplies all 16 fields.
58
+
59
+ After exhausted startup attempts or an analysis deadline, explicit user authorization may record a one-fingerprint `review-runtime-override` with identity, reference, reason, evidence, and `--confirm-runtime-bypass`.
60
+ It records `runtime-bypassed`, never `approved`, and cannot replace a substantive `blocked` or `changes-requested` result.
62
61
 
63
62
  ## Close with the bounded routed gate
64
63
 
@@ -95,7 +94,7 @@ python3 docs/tools/autopipeline/ap.py doctor
95
94
  ```
96
95
 
97
96
  Finish registered tasks before changing versions. `autocoding init` idempotently
98
- replaces managed constraints, preserves schema-supported values, installs canonical
99
- docs/templates, archives obsolete content outside active `docs/`, and preserves
100
- designated artifacts. It verifies the managed-install manifest; `doctor` cheaply
101
- rechecks declared paths without treating project-owned files or archives as a mirror.
97
+ replaces managed defaults and constraints while preserving the project overlay
98
+ byte-for-byte after its one-time semantic migration. It installs canonical docs,
99
+ archives obsolete content outside active `docs/`, verifies the managed-install
100
+ manifest, and lets `doctor` validate the effective configuration.
@@ -1,8 +1,8 @@
1
1
  {
2
2
  "schema_version": 1,
3
3
  "managed_versions": {
4
- "agents": "4.2.7",
5
- "engineering": "4.2.7"
4
+ "agents": "4.3.0",
5
+ "engineering": "4.3.0"
6
6
  },
7
7
  "known_official_agents_sha256": [
8
8
  "ebaaadab60327b3a4852948443d7191a4d48cb00bf4910038ab78c9149089373"
@@ -1,6 +1,6 @@
1
1
  ---
2
2
  workflow:
3
- skill_version: "4.2.7"
3
+ skill_version: "4.3.0"
4
4
  mode: "dev"
5
5
  profile: "auto"
6
6
  completion: "push"
@@ -83,10 +83,12 @@ docs:
83
83
  api_change_log: "docs/interfaces/api-change-log.md"
84
84
  ---
85
85
  # Engineering Configuration and Project Facts
86
- <!-- auto-coding-skill:managed-workflow:start version=4.2.7 -->
86
+ <!-- auto-coding-skill:managed-workflow:start version=4.3.0 -->
87
87
  Root `AGENTS.md` is the single shared behavioral protocol. The installed
88
- `SKILL.md` contains invocation guidance. This file is the exact-schema source for
89
- project workflow configuration and access values that neither file can infer.
88
+ `SKILL.md` contains invocation guidance. This whole file is the managed default
89
+ configuration contract. Project-owned values live only in
90
+ `docs/project/auto-coding-skill.yaml`; the runtime overlays that file on these
91
+ defaults without allowing it to weaken code-enforced safety invariants.
90
92
  The frontmatter contract is:
91
93
  - `workflow`: installed version and push-completion mode.
92
94
  - `access`: required project, Jenkins, GitLab, and Nexus access values.
@@ -95,16 +97,19 @@ The frontmatter contract is:
95
97
  - `validation`: explicit path-to-command routes plus the bounded final-gate time
96
98
  budget. Code/config paths must be mapped; route-level `timeout_seconds` can
97
99
  bound a slower command; commands inherit PATH and declare intentional activation.
98
- - `risk`: project-specific signals that increase design/review depth but never
99
- expand the automatic local gate.
100
- - `structure` and `optimization`: project-selected enforcement and completion policy; frontmatter values are authoritative.
100
+ - `risk`: default signals that project-specific rules may replace in the
101
+ project overlay; code-enforced high-risk classifications remain mandatory.
102
+ - `structure` and `optimization`: managed defaults that the project overlay may
103
+ replace explicitly.
101
104
  - `docs`: locations inside the managed documentation directory framework.
102
105
  Run `python3 docs/tools/autopipeline/ap.py doctor` after changing frontmatter; it
103
106
  also performs the bounded local managed-install integrity check.
104
- Do not put competing delivery flows, agent roles, lifecycle rules, gate semantics,
105
- or completion rules outside this managed block.
107
+ Do not edit this file for project specialization. Run
108
+ `python3 docs/tools/autopipeline/ap.py config-effective --json` to inspect the
109
+ redacted effective configuration, and edit only the project overlay. Do not put
110
+ competing delivery flows, agent roles, lifecycle rules, gate semantics, or
111
+ completion rules outside this managed block.
106
112
  <!-- auto-coding-skill:managed-workflow:end -->
107
113
  # Project Facts Location
108
- Keep durable project facts in the installed `docs/project/` files. Do not append
109
- workflow rules or additional sections here; initialization owns this document's
110
- schema and body.
114
+ Keep durable project facts in the installed `docs/project/` files. Initialization
115
+ owns this document byte-for-byte and never rewrites an existing project overlay.
@@ -1,8 +1,8 @@
1
- <!-- auto-coding-skill:managed-agents:start version=4.2.7 -->
1
+ <!-- auto-coding-skill:managed-agents:start version=4.3.0 -->
2
2
  # Shared Engineering Protocol
3
3
 
4
- This file is fully managed by `auto-coding-skill`. Keep project-specific facts,
5
- access values, risk rules, and validation routes in `docs/ENGINEERING.md`.
4
+ This file is fully managed by `auto-coding-skill`. Keep project-specific facts in
5
+ `docs/project/` and configuration in `docs/project/auto-coding-skill.yaml`.
6
6
 
7
7
  ## Authority
8
8
 
@@ -11,8 +11,8 @@ Use this order when sources disagree:
11
11
  1. The current user request.
12
12
  2. Code, tests, schemas, migrations, and runtime configuration for behavior that
13
13
  exists now.
14
- 3. `docs/ENGINEERING.md` for workflow configuration and access values, then the
15
- managed `docs/project/` files for durable project facts.
14
+ 3. The effective configuration formed from managed `docs/ENGINEERING.md` defaults
15
+ plus the project overlay, then `docs/project/` for durable project facts.
16
16
  4. Current interface documentation for intended contracts.
17
17
  5. Accepted DD/ADR for lasting decisions.
18
18
  6. Taskbooks, closures, reviews, and deployment records as history only.
@@ -79,22 +79,19 @@ failure may continue in the same conversation/task without a second lifecycle.
79
79
  - Validate delegated assignment/result JSON with `agent-contract-check` before use.
80
80
  - Require fingerprinted review only for path/rule-confirmed high-risk,
81
81
  contract-crossing, parallel, or configured work; intent words are candidates only.
82
- - Keep `reviewer` at `xhigh`: focused is one 90-second analysis; deep gets 300
83
- seconds for parallel/cross-module or sensitive boundaries. Timeout is `blocked`
84
- and is not sent to `task-review`.
82
+ - Use `high`/150 seconds for focused review and `xhigh`/360 seconds for deep parallel,
83
+ cross-module, or sensitive review. Retry once only if no semantic event arrives
84
+ within 30 seconds; never retry started analysis.
85
85
  - Run `python3 docs/tools/autopipeline/ap.py review-run <TASK> --reviewer <ID> --json`.
86
- It launches a separate read-only Codex process without the lifecycle-owner identity, stops its process group at the fixed deadline,
87
- and records only the assigned HEAD, scope, identity, fingerprint, and mode-0600 SHA-256-bound Git-local diff artifact.
88
- Before analysis, the Reviewer runs `python3 docs/tools/autopipeline/ap.py review-artifact --file <assignment.json>` and reviews that frozen patch, never a reconstructed live `git diff`. `review-assignment` alone requires a different
89
- deadline-capable host and cannot stop an in-app subagent.
90
- - `agent-result-template` supplies all 16 fields; contract checking aggregates errors,
91
- and `review-run` safely normalizes presentation fields from the same analysis.
92
- - Review blocks only defects introduced or worsened in the promised scope.
93
- Adjacent existing issues are non-blocking follow-ups.
94
- - Semantic changes invalidate approval. Mechanical documentation-only corrections
95
- may receive a targeted recheck by the same reviewer.
96
- - Historical debt does not block ordinary delivery; block only new or worsened
97
- P0/P1 issues.
86
+ It runs a separate read-only identity, streams allowlisted metadata privately,
87
+ stops the process group at deadline, and binds a mode-0600 SHA-256 patch to the
88
+ assigned HEAD, scope, identity, and fingerprint. Reviewers read the frozen patch
89
+ with `review-artifact`; `review-assignment` alone needs a deadline-capable host and cannot stop an in-app subagent.
90
+ - `agent-result-template` supplies all 16 fields and `review-run` normalizes presentation.
91
+ - Runtime exhaustion needs explicit `review-runtime-override` authorization; it records
92
+ `runtime-bypassed`, never approval, and cannot replace a substantive blocked result.
93
+ - Block only new or worsened in-scope P0/P1 defects. Adjacent debt is a follow-up;
94
+ semantic changes invalidate approval, while mechanical docs may get a targeted recheck.
98
95
 
99
96
  ## Bounded real validation
100
97
 
@@ -127,9 +124,9 @@ failure may continue in the same conversation/task without a second lifecycle.
127
124
  ## Documentation and access
128
125
 
129
126
  - This AGENTS file is the single behavioral protocol. The installed `SKILL.md`
130
- contains invocation guidance. `docs/ENGINEERING.md` is the exact project
131
- configuration source; `docs/project/` owns durable facts. Do not repeat the
132
- protocol elsewhere.
127
+ contains invocation guidance. `docs/ENGINEERING.md` is managed default policy;
128
+ `docs/project/auto-coding-skill.yaml` is the higher-priority project overlay,
129
+ and the remaining `docs/project/` files own durable facts.
133
130
  - `autocoding init` owns the active `docs/` directory topology and managed
134
131
  templates. Put durable facts into the installed project files. Architecture,
135
132
  ADR, interface, DD, review, and deploy-record artifacts may use their designated
@@ -139,6 +136,10 @@ failure may continue in the same conversation/task without a second lifecycle.
139
136
  - Ordinary work creates no taskbook, closure Markdown, evidence JSONL, active-task
140
137
  document, or design file. Machine coordination/evidence stays in Git
141
138
  common/local state and cannot change the reviewed diff.
139
+ - After diagnosing the project/Skill boundary, record a probable shared-Skill
140
+ defect or gap from `docs/skill-feedback/_TEMPLATE-SKILL-FEEDBACK.md`; keep
141
+ project-only configuration out, redact sensitive/raw evidence, and never mutate
142
+ a frozen Reviewer diff merely to add feedback.
142
143
  - Pure ledger/archive reconciliation is terminal: validate, commit once, and stop.
143
144
  - Fill configured project/Jenkins/GitLab/Nexus access fields during initialization.
144
145
  Plaintext values are allowed; do not invent or unnecessarily echo credentials.