Skip to content

Commit ca23241

Browse files
feat(skill): 1.136.0 — wave 57 de-duplicate the domain-checklists middle tier (#331)
#325 Option A: 3 copied definitions -> gist + canonical-deep-file pointer (do-no-harm verified, zero info lost); 2 quick-scan lists kept. Trio -> 1.136.0. Closes #325. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
1 parent c9611eb commit ca23241

10 files changed

Lines changed: 39 additions & 34 deletions

File tree

‎.claude-plugin/plugin.json‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@
22
"name": "deep-code-review",
33
"description": "Perun — bring the thunder to your codebase. A portable, evidence-grounded quality bar for building with AI agents: the flagship deep-code-review skill (default install is review-only), plus opt-in overlays — gated delivery, an idea critic, a suite conductor, and stage-aware product, growth, positioning, business-ops, and output-safety specialists.",
44
"skills": "./.claude/skills",
5-
"version": "1.135.0",
5+
"version": "1.136.0",
66
"author": {
77
"name": "deep-code-review contributors"
88
},

‎.claude/skills/agentic-delivery/SKILL.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@ description: >-
1010
license: MIT
1111
metadata:
1212
author: deep-code-review contributors
13-
version: "1.135.0"
13+
version: "1.136.0"
1414
---
1515

1616
# Agentic delivery
Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
1.135.0
1+
1.136.0

‎.claude/skills/deep-code-review/SKILL.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -11,7 +11,7 @@ description: >-
1111
license: MIT
1212
metadata:
1313
author: deep-code-review contributors
14-
version: "1.135.0"
14+
version: "1.136.0"
1515
---
1616

1717
# Deep Code Review
Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
1.135.0
1+
1.136.0

‎.claude/skills/deep-code-review/references/domain-checklists.md‎

Lines changed: 16 additions & 20 deletions
Original file line numberDiff line numberDiff line change
@@ -106,9 +106,9 @@ unwalked domain.
106106
boundary **and** the confidentiality tier (S0–S3 below) (`git check-ignore`,
107107
`git ls-files --error-unmatch`, route enumeration + anonymous GET sweep), not
108108
to the rendering code. And an `Origin`/`Referer`/`Sec-Fetch-Site` check is
109-
**CSRF defense, not authentication** — bypassable by any non-browser client and
110-
absent on many GET navigations; if it is the only gate on a sensitive/paid/
111-
mutating action, that action is effectively unauthenticated.
109+
**CSRF defense, not authentication** — if it is the only gate on a sensitive/paid/
110+
mutating action, that action is effectively unauthenticated (why it is bypassable,
111+
and the CWE-352-vs-CWE-306 split, are in `security-appsec.md`).
112112
- **Secrets**: nothing sensitive in source, history, comments, logs, error
113113
strings, or fixtures; env/secret-manager only. A gate that finds a secret
114114
reports `file:line` — it **never echoes the secret**. User-facing errors expose
@@ -167,18 +167,14 @@ skill**, also walk AST01–AST10 in `references/security-agent-skills.md`.
167167
### D. Data integrity & data quality → `references/data-quality.md`
168168
Apply to any pipeline, ETL, enrichment, scraping, or dataset producer. Judge the
169169
**output**, not just the code.
170-
- **Monotonic quality (hard invariant)**: a write/merge may **never** replace a
171-
populated, higher-confidence value with an empty, lower, or duplicate one.
172-
Upserts **field-merge with preserve-if-absent** — never wholesale-replace.
173-
Require a regression test that fails on this exact mode. Non-regression gate:
174-
a within-dataset uniqueness check **plus** a populated→worse check against an
175-
**explicitly pinned** baseline (never the live artifact).
176-
- A **degraded/empty/fallback result surfaced by a "latest/max" read is the same
177-
monotonic breach** even when no overwrite occurred — tag it (`degraded: true`)
178-
and have latest/best queries skip it. A write/erosion guard must intercept
179-
**every** mutation primitive (update AND clear AND append AND delete), not just
180-
the common one, and the test proving its coverage must **discover** write-sites
181-
(grep/AST), never hardcode a list that rots (see `references/data-quality.md`).
170+
- **Monotonic quality (hard invariant)**: a write/merge never replaces a
171+
populated, higher-confidence value with an empty/lower/duplicate one — upserts
172+
**field-merge with preserve-if-absent**, and a degraded/fallback surfaced by a
173+
**"latest/max" read is the same breach** even when nothing was overwritten. The
174+
full invariant and the two-part non-regression gate (within-dataset uniqueness
175+
**plus** a populated→worse check against a pinned baseline) live in
176+
`references/data-quality.md` §1; the write-erosion guard's every-mutation-primitive
177+
coverage + discover-write-sites test are in its §5 — walk it for any data producer.
182178
- **No fabrication in the data**: skip a field rather than guess; corroboration =
183179
**two+ independent sources**; `inferred` ≠ `sourced`; omit the unverifiable.
184180
- **Entity resolution biases false-exclude over false-merge**: stable-id/proof
@@ -517,11 +513,11 @@ export-boundary suppression, and licence-compatibility detail live there.
517513

518514
### S. Branches, merges & open-work triage → `references/branch-and-merge-hygiene.md`
519515
Apply on a **FULL / repo-level review**, or whenever the request names branches,
520-
cleanup, or open work. **N/A by scope on a narrow `DIFF`/`FILE`** — a PR/branch
521-
reviewed against a base stays a compact packet (don't fetch and triage every
522-
branch to review a ten-line change) unless branch cleanup was explicitly asked.
523-
The deliverable is a **triage of all open work**: for every branch, one
524-
recommendation and the exact command. Distinct from section O, which owns whether
516+
cleanup, or open work; **N/A by scope on a narrow `DIFF`/`FILE`**. The scope rule
517+
(a base-reviewed PR/branch stays a compact packet — don't triage every branch to
518+
review a ten-line change) and the deliverable (a **triage of all open work**: one
519+
recommendation + the exact command per branch) are owned by
520+
`references/branch-and-merge-hygiene.md`. Distinct from section O, which owns whether
525521
branch *protection* is configured — this owns *what open work exists and what to
526522
do with it*; the one seam ("must this merge go through a PR?") reads O's posture.
527523
- **Ground the branch set before judging it** — `git fetch --all --prune` first;

‎.claude/skills/idea-critic/SKILL.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@ description: >-
99
license: MIT
1010
metadata:
1111
author: deep-code-review contributors
12-
version: "1.135.0"
12+
version: "1.136.0"
1313
---
1414

1515
# Idea critic

‎.claude/skills/idea-critic/VERSION‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
1.135.0
1+
1.136.0

‎CHANGELOG.md‎

Lines changed: 9 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -3,6 +3,15 @@
33
All notable changes to this repository are documented here. Format loosely
44
follows Keep a Changelog; versioning follows Semantic Versioning.
55

6+
## [1.136.0] — 2026-09-18
7+
8+
Wave 57 — `domain-checklists.md` middle-tier de-duplication, **#325 Option A** (owner-selected). The checklist tier restated definitions its per-domain deep files own — the repo's own anti-duplication thesis, violated. Surgically converted the **3 copied definitions/rules the #325 audit surfaced** to a scannable check + a pointer to the canonical deep file (each verified do-no-harm: the full detail was confirmed present in the deep file before the copy was dropped):
9+
- **Monotonic-quality invariant** (domain D) — near-verbatim from `data-quality.md` (dc even linked while copying); now gist + `data-quality.md` §1 (full invariant + two-part non-regression gate) and §5 (the every-mutation-primitive / discover-write-sites coverage).
10+
- **CSRF-guard-is-not-authentication** (domain B) — the "why it's bypassable" + the CWE-352-vs-CWE-306 split now point to `security-appsec.md`; the check (CSRF ≠ auth → effectively unauthenticated) stays.
11+
- **Open-work triage scope + deliverable** (domain S) — a 3rd verbatim copy; the compact-packet scope rule + the triage deliverable now point to `branch-and-merge-hygiene.md`; the domain-S-unique O/S seam stays.
12+
13+
The other two audit pairs (the F and I sections' 🚩 quick-scan lists) were **assessed and kept**: a 🚩 quick-scan IS the checklist's core function (dc:7), the section headers already route to the fuller lists in the deep files, and dropping them would regress the reviewer's scannability — do-no-harm over literal conversion (A1 rated them Nit/acceptable). Trio → 1.136.0.
14+
615
## [1.135.0] — 2026-09-18
716

817
Wave 56 — link-rot / citation-freshness audit of `docs/standards-index.md` (the "verify before citing" ledger, which `install.sh` vendors into every install's `references/`). Re-fetched every cited URL: content was accurate (**zero drift across ~90 URLs**) but not fully live — **1 dead + 4 moved**, now fixed and each re-verified by direct fetch this session:

‎SHA256SUMS‎

Lines changed: 7 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
11
0243c4d5712d24c5f166143bb37da9407b96fb61005ba36993d4bc3766b90a29 .claude/skills/agentic-ceo/SKILL.md
22
cc52f678848b814373757b460383bf61960e4943c203735adde0a350b3e50989 .claude/skills/agentic-ceo/VERSION
33
3122784e8b188babadd7ceb0dddf69cfb308e7a02bbf67ed82cf3527a81f5eb1 .claude/skills/agentic-ceo/evals/evals.json
4-
3f652e64e7c5b13138f5f3deac589bee8e4626ebdc3bdd16ce64dff2943fd9af .claude/skills/agentic-delivery/SKILL.md
5-
35964059adf4af03ce64f60373aa100bcb161535a9db80654332273474488d13 .claude/skills/agentic-delivery/VERSION
4+
9b48f58c7fac67086ab999b4d3d59e18ce9a0e048193d5594ec153c82efee786 .claude/skills/agentic-delivery/SKILL.md
5+
0a6e0a34a5d02777e09975b15c3376b07643f6cfe7a15990f6842b70eff696bb .claude/skills/agentic-delivery/VERSION
66
8bf00d58a064d81bc6f28f919d5b6f7d8c174c8317496f77c43bc247908c5c4a .claude/skills/agentic-delivery/evals/evals.json
77
863b7535b90efad3c661a8eb3a24fc6356c6fa54e8874c05bee11f110d30274b .claude/skills/agentic-delivery/references/fast-agentic-delivery.md
88
200e58588607d6c444811d91722cc327bbb6b9a5236dcd0c03d3d70936c61b30 .claude/skills/agentic-delivery/references/host-enforcement.md
@@ -25,8 +25,8 @@ cc52f678848b814373757b460383bf61960e4943c203735adde0a350b3e50989 .claude/skills
2525
359e11a6b81b9589cdc0d880c175b2876410d58a2cf22fb8c3e06e075b67e01d .claude/skills/contribution/evals/evals.json
2626
3f047b5935d72317605691d0fc4e6357cf6c44d1b91347eb2fc36e75497f7382 .claude/skills/contribution/kernel-paths.txt
2727
bb541f96bd7d0c3bb4adff8dcaac0f50efa2493b8215b6aa671805bfbb89c42b .claude/skills/contribution/references/contribution-procedure.md
28-
766fe65d3b5d768a35dcbe41085d916c0b8fe58f2563290d077f9ccacc90c343 .claude/skills/deep-code-review/SKILL.md
29-
35964059adf4af03ce64f60373aa100bcb161535a9db80654332273474488d13 .claude/skills/deep-code-review/VERSION
28+
f39f88b45bbf6a6227f9b53ffe67f8718bf760a89c4d9972eda8d18e94d78458 .claude/skills/deep-code-review/SKILL.md
29+
0a6e0a34a5d02777e09975b15c3376b07643f6cfe7a15990f6842b70eff696bb .claude/skills/deep-code-review/VERSION
3030
44196b30b6fe480289d829ac6e9f044de939bfea7a73d064e72dc448509aab2a .claude/skills/deep-code-review/evals/evals.json
3131
960211dc69f2b40ba80cc0ad1068816328f1c2dbd465b552b51ae5c805e7fe75 .claude/skills/deep-code-review/references/api-contracts.md
3232
93d695b99b448a59b1b7e4de3449692847089ba36e9530d0e24d19c04650efc1 .claude/skills/deep-code-review/references/billing-correctness.md
@@ -36,7 +36,7 @@ e969296002970b936576e3eda49a319826404534af11120984eba5d08c639505 .claude/skills
3636
b8a8036bc431ea26503895d8c20f6751500bd38bd2edaccf602825f028bc3fd3 .claude/skills/deep-code-review/references/dependency-currency-and-upgrades.md
3737
b03c0b7dd595cfcfdd52b258ffb313816fdc6f8e1d2a814404690631745049df .claude/skills/deep-code-review/references/docs-and-dx.md
3838
b0bbec43934a3da9e666183dc157c97ef7e587670efe26209d12a0e0f7f01900 .claude/skills/deep-code-review/references/docs-evolution-by-stage.md
39-
41c447145ef838ea4fe979dfcb8b279fa0f6728738c63481fe779b62810d3320 .claude/skills/deep-code-review/references/domain-checklists.md
39+
ee738330b78847d5ee1bb0c5b74be9b4a1c0ce1292bf980e3e14b0c59ada7de0 .claude/skills/deep-code-review/references/domain-checklists.md
4040
ed10ed59f4ec9fcfb337ce474a8971b6a4a17f78bd0980db9a6bed08a308166f .claude/skills/deep-code-review/references/example-review-report.md
4141
e82bd81a3cce71fb43c8e2d22b5c418b2895d158c6505f6f7d7f1157a752ddc4 .claude/skills/deep-code-review/references/frontend-a11y.md
4242
d08c079af97d48bb9dc4593ee116a01b848a995ba2845c08d0e5dd23b60e509b .claude/skills/deep-code-review/references/infra-evolution-by-stage.md
@@ -67,8 +67,8 @@ a0ed7ad061b7e60230601ee9b38a00400841d6ddf3af386edbd4b34a981c9798 .claude/skills
6767
5068e60f9615684c8d8bf8fbd7896748267cf6688895a701bdd25cd945181cb8 .claude/skills/growth-analytics/SKILL.md
6868
cc52f678848b814373757b460383bf61960e4943c203735adde0a350b3e50989 .claude/skills/growth-analytics/VERSION
6969
4c8047eeb8b174b520c736e7b189d5a49e4a755aad738e9c44c8af4e25fe1067 .claude/skills/growth-analytics/evals/evals.json
70-
ef001dc75d5ea517f95a49d04a12c8c32a756780db5d9ac284d4f5afdf8cfc18 .claude/skills/idea-critic/SKILL.md
71-
35964059adf4af03ce64f60373aa100bcb161535a9db80654332273474488d13 .claude/skills/idea-critic/VERSION
70+
7811a008ac1ef83214b1547a6815ec309a6868263ddbf4cb753825894c167a93 .claude/skills/idea-critic/SKILL.md
71+
0a6e0a34a5d02777e09975b15c3376b07643f6cfe7a15990f6842b70eff696bb .claude/skills/idea-critic/VERSION
7272
c59f0b07326a2bcf7a5bfcd3286059c31d77d6f359aa615d7990e579b2f52f9f .claude/skills/idea-critic/evals/evals.json
7373
c5615e5c59bc2e452c2c49619be2ec5e456f3c167e772498316f8b5a45deb2c8 .claude/skills/idea-critic/references/decision-hygiene.md
7474
612a8b0426f4cad71f88aa5aead935351a6c586b74bf34d8ef39f2df2f9f72d1 .claude/skills/idea-critic/scripts/validate_verdict.py

0 commit comments

Comments
 (0)