devrites 3.0.5 → 3.0.6

package/pack/generated/codex/skills/rite-vet/reference/review-axes.md

@@ -1,25 +1,25 @@
-# The vet review — scope challenge, four axes, required outputs
+# The vet review: scope challenge, four axes, required outputs
 
 The body of `$rite-vet`. Run §0 first as a blocking gate, then the four axes one at a time,
 then the required outputs. Apply the senior-engineer lenses in [`eng-lenses.md`](eng-lenses.md)
-throughout — they're how you *see* the findings, not a separate checklist. Every finding is
+throughout. They're how you *see* the findings, not a separate checklist. Every finding is
 calibrated and gated (see "Confidence + verification gate" below) before it reaches the human.
 
 ---
 
-## §0. Scope Challenge (blocking gate — runs before any axis)
+## §0. Scope Challenge (blocking gate: runs before any axis)
 
 Before reviewing *how* the plan builds, challenge *whether it should build this much*. This is
-implementation-scope discipline (the spec's ambition is settled — that was `$rite-temper`).
+implementation-scope discipline (the spec's ambition is settled, that was `$rite-temper`).
 
 1. **What already exists?** For each sub-problem in the plan, find the existing code/flow that
-   already solves it (use a code-intelligence index if available — see `../../devrites-lib/reference/standards/tooling.md`).
+   already solves it (use a code-intelligence index if available: see `../../devrites-lib/reference/standards/tooling.md`).
    Can the plan **capture outputs from an existing
    flow** instead of building a parallel one? Reuse → extend → build new, in that order
    (`coding-style.md`). List every reuse opportunity the plan misses.
 2. **Minimum diff.** What's the smallest set of changes that meets the spec's *acceptance
    criteria*? Flag any planned work that can be deferred without blocking acceptance. Be ruthless
-   about implementation scope creep — but never cut an acceptance criterion (that's a Drift Guard
+   about implementation scope creep, but never cut an acceptance criterion (that's a Drift Guard
    matter, not a trim).
 3. **Complexity smell.** If `plan.md` touches **>8 files** or adds **>2 new services / modules /
    classes**, treat it as a smell. Check the plan's complexity gate justifies it. If it doesn't
@@ -27,17 +27,17 @@ implementation-scope discipline (the spec's ambition is settled — that was `$r
    via `AskUserQuestion` whether to reduce or proceed. Do not start the axes until answered.
 4. **Built-in check.** For each new pattern / infra component / concurrency approach the plan
    introduces, verify a framework/runtime built-in doesn't already do it, and that the choice is
-   current best practice with no known footgun — dispatch `devrites-source-driven` to confirm at
+   current best practice with no known footgun: dispatch `devrites-source-driven` to confirm at
    the source and record the citation. A custom roll where a built-in exists is a scope-reduction
    finding.
 5. **Completeness check.** Is the plan doing the complete version or a shortcut? With AI-assisted
    coding the cost of completeness (full edge-case handling, complete error paths, real test
-   coverage) is a fraction of what it was — a shortcut that saves human-hours but only saves
+   coverage) is a fraction of what it was: a shortcut that saves human-hours but only saves
    minutes here is a false economy. Prefer the complete option; flag shortcuts that exist only to
    save effort that AI has already made cheap.
 6. **Distribution check.** If the plan introduces a new artifact (CLI binary, package, container,
    deployable), does it include how it gets built / published / installed? If distribution is
-   deferred, say so explicitly in "NOT in scope" — don't let it silently drop.
+   deferred, say so explicitly in "NOT in scope": don't let it silently drop.
 
 > **STOP discipline.** If the complexity smell trips, the `AskUserQuestion` is a tool call, not
 > prose. Naming the 80%-solution in chat and continuing is the failure this gate exists to prevent.
@@ -49,14 +49,14 @@ If the smell does not trip, present the §0 findings and proceed to Axis 1.
 ## The four axes (one at a time, ≤8 findings each)
 
 For each axis: evaluate, then **walk each finding WITH the human** via `AskUserQuestion` (one
-issue per call — see "How to ask" below). HITL pauses on each material finding; AFK auto-applies
+issue per call: see "How to ask" below). HITL pauses on each material finding; AFK auto-applies
 within the gate ceiling (`depth.md`). If an axis genuinely has no issue, say "No issues,
-moving on" and continue — don't manufacture findings.
+moving on" and continue: don't manufacture findings.
 
 ### 1. Architecture
 - Component boundaries, coupling, data-flow patterns, single points of failure. Architecture records invariants, not scaffolding: each medium+ decision should state `Binds:` and `Prevents:` so the builder knows what divergence it prevents.
 - Scaling characteristics; where the plan's approach breaks under real load.
-- Security architecture at the seams (auth, data access, API boundaries) — does the plan name
+- Security architecture at the seams (auth, data access, API boundaries): does the plan name
   the trust boundary for each untrusted input?
 - For each new codepath / integration point: **one realistic production failure scenario** and
   whether the plan accounts for it (feeds the failure-mode table).
@@ -65,7 +65,7 @@ moving on" and continue — don't manufacture findings.
 
 ### 2. Plan code-quality
 - Module structure the plan implies; DRY across the slices (flag planned repetition aggressively).
-- Error-handling + edge cases the plan names — and the ones it doesn't (call those out explicitly).
+- Error-handling + edge cases the plan names, and the ones it doesn't (call those out explicitly).
 - Over-engineering (premature abstraction, an extension point with no second caller) vs
   under-engineering (fragile / hacky) relative to `patterns.md` + `coding-style.md`.
 - Tech-debt hotspots the plan walks into; existing inline diagrams in touched files that the
@@ -73,21 +73,21 @@ moving on" and continue — don't manufacture findings.
 
 ### 3. Test-coverage design
 The differentiator: design the tests *before* the code, so the build writes them alongside.
-- **Framework detection** — find the project's existing test runner + conventions; match them
-  (never introduce a new runner to prove one change — `testing.md`).
+- **Framework detection:** find the project's existing test runner + conventions; match them
+  (never introduce a new runner to prove one change: `testing.md`).
 - **Map acceptance → tests.** Every spec acceptance criterion must map to ≥1 planned, surface-anchored test (the API response/UI state/CLI output the criterion names, not an internal proxy).
-- **Tool per path** — unit (pure logic, single function, edge cases), integration/E2E (a user
+- **Tool per path:** unit (pure logic, single function, edge cases), integration/E2E (a user
   flow spanning 3+ components, an auth/payment/data-loss path, a mock-hides-failure boundary),
   eval (an LLM/prompt change that needs a quality bar).
-- **Interaction inventory (UI slices) — enumerate every interactive element + flow.** List each
+- **Interaction inventory (UI slices): enumerate every interactive element + flow.** List each
   input field, checkbox, radio, select, toggle, button, and actionable link, plus each user
-  flow; assign each ≥1 asserting test **at the right level** — elements/fields → unit/component,
+  flow; assign each ≥1 asserting test **at the right level**: elements/fields → unit/component,
   critical journeys → one E2E (never one-per-field). Every element/flow with no asserting test
   is a GAP; no element ships unverified. Write the inventory to `test-plan.md` (table in
   `artifacts.md`). This is `testing.md` "Completeness" made concrete for the plan.
 - **Regression rule (mandatory, no question):** when the plan modifies existing behavior and the
   current suite doesn't cover the changed path, a regression test is added to the plan as a
-  **Critical** requirement — no `AskUserQuestion`, no skipping. Regressions are the highest-priority
+  **Critical** requirement: no `AskUserQuestion`, no skipping. Regressions are the highest-priority
   test because they prove something broke. When unsure whether a change is a regression, write the test.
 - Produce the **coverage diagram** (shape below) and add a specific test requirement per GAP.
 
@@ -112,7 +112,7 @@ acceptance criteria covered ✓" and continue.
 ### 4. Performance
 - N+1 / unbounded queries and DB access patterns; memory concerns; caching opportunities;
   high-complexity hot paths.
-- Per `performance.md`: name a number to measure or a budget — don't recommend speculative
+- Per `performance.md`: name a number to measure or a budget: don't recommend speculative
   micro-tuning. A perf finding is "measure X against budget Y", not "this feels slow".
 
 ---
@@ -126,7 +126,7 @@ Tag each finding `[severity] (confidence: N/10) <plan/task/spec ref> — finding
 **The gate:** before raising a finding, quote the line(s) that motivate it. Can't quote it →
 force confidence ≤4 and suppress. This kills the "the plan doesn't handle X" finding when the
 plan *does* and you skimmed. Don't fabricate 7+ to dodge it. (Same discipline the reviewer agent
-runs — `devrites-plan-reviewer`.)
+runs: `devrites-plan-reviewer`.)
 
 ---
 
@@ -144,24 +144,24 @@ Use `AskUserQuestion` per the pack's standard. Plan-review specifics:
   happy-path), add `Completeness: N/10` per option. If they differ in *kind* (two different
   architectures), skip the score and note "options differ in kind, not coverage". Never fabricate
   a score on a kind question.
-- Every material finding ends as a **recorded decision** — a resolved `questions.md` qid (HITL) or
-  a `decisions.md` ADR (AFK) — so the walk leaves an auditable trail, not just chat.
+- Every material finding ends as a **recorded decision**: a resolved `questions.md` qid for HITL
+  or a `decisions.md` ADR for AFK. The walk must leave an auditable trail, not just chat.
 
 ---
 
 ## Required outputs (after the axes)
-1. **"NOT in scope"** — work considered and explicitly deferred, one-line rationale each. Folds
+1. **"NOT in scope":** work considered and explicitly deferred, one-line rationale each. Folds
    into `plan.md` §Scope boundaries + `spec.md` Non-goals (via the Guard) so it can't silently re-enter.
-2. **"What already exists"** — existing code/flows that solve sub-problems, and whether the plan
+2. **"What already exists":** existing code/flows that solve sub-problems, and whether the plan
    reuses or rebuilds them. Every missed reuse becomes a §0 finding.
-3. **Failure-mode table** — for each new codepath: a realistic failure, and whether (a) a test
+3. **Failure-mode table:** for each new codepath: a realistic failure, and whether (a) a test
    covers it, (b) error handling exists, (c) the user sees a clear error or a silent failure. A
    failure with **no test AND no handling AND silent** is a **Critical gap**. (Shape in
    [`artifacts.md`](artifacts.md).)
-4. **Worktree parallelization strategy** — analyze the slices for parallel execution. **Skip** if
+4. **Worktree parallelization strategy:** analyze the slices for parallel execution. **Skip** if
    all slices touch the same module or there are <2 independent workstreams ("Sequential, no
    parallelization opportunity"). Otherwise: a dependency table (module-level, not file-level),
    parallel lanes (shared module → same lane/sequential; independent → separate lanes), execution
    order, and conflict flags where two lanes touch the same module dir. This feeds `$rite-build`'s
    isolation strategy and the autocomplete loop. (Shape in [`artifacts.md`](artifacts.md).)
-5. **Completion summary** — the one-glance recap (shape in [`artifacts.md`](artifacts.md)).
+5. **Completion summary:** the one-glance recap (shape in [`artifacts.md`](artifacts.md)).

package/pack/generated/codex/skills/rite-zoom-out/SKILL.md

@@ -12,61 +12,61 @@ This is the Codex mirror of a DevRites skill. In Codex:
 
 - Load DevRites engineering standards from `.agents/skills/devrites-lib/reference/standards/`. Read `.agents/skills/devrites-lib/reference/standards/core.md` before workflow work, then load the other `.agents/skills/devrites-lib/reference/standards/*.md` files exactly when this skill asks for them.
 - Use the installed `devrites-engine` binary as the canonical runtime helper surface for orientation, gates, and state mutation.
-- When this skill asks for a DevRites specialist or writer agent, **explicitly** spawn the matching Codex custom agent from `.codex/agents/devrites-*.toml` through Codex subagents (`spawn_agent`), then wait for its result and reconcile it as the skill instructs. Do not do the review inline just because the instruction to spawn is embedded here — Codex under-fires embedded spawn/skill instructions (openai/codex #23496), so treat the spawn as required, not optional.
-- The independence of a fresh-context subagent is the point. If Codex genuinely cannot spawn subagents in the current surface, run the documented inline fallback and **label the result an inline fallback, not an independent review** — an inline pass shares the calling context and is weaker evidence.
+- When this skill asks for a DevRites specialist or writer agent, **explicitly** spawn the matching Codex custom agent from `.codex/agents/devrites-*.toml` through Codex subagents (`spawn_agent`), then wait for its result and reconcile it as the skill instructs. Do not do the review inline just because the instruction to spawn is embedded here: Codex under-fires embedded spawn/skill instructions (openai/codex #23496), so treat the spawn as required, not optional.
+- The independence of a fresh-context subagent is the point. If Codex genuinely cannot spawn subagents in the current surface, run the documented inline fallback and **label the result an inline fallback, not an independent review**: an inline pass shares the calling context and is weaker evidence.
 - Codex project hooks are installed in `.codex/hooks.json`. Review and trust them with `/hooks` before relying on hook enforcement.
-- When this skill asks a HITL question via `AskUserQuestion`: Codex's equivalent (`request_user_input`) exists only in Plan mode. Outside Plan mode, render the option set as a plain numbered list in chat and **end the turn** so the human answers — NEVER silently pick an option yourself; auto-picking is AFK's contract, gated by the `.devrites/AFK` sentinel.
+- When this skill asks a HITL question via `AskUserQuestion`: Codex's equivalent (`request_user_input`) exists only in Plan mode. Outside Plan mode, render the option set as a plain numbered list in chat and **end the turn** so the human answers: NEVER silently pick an option yourself; auto-picking is AFK's contract, gated by the `.devrites/AFK` sentinel.
 
 
-# $rite-zoom-out — step up one abstraction layer
+# $rite-zoom-out: step up one abstraction layer
 
 When the agent (or the user) is staring at unfamiliar code without a working mental
 model of how it fits the larger system. Stops the "open more files" reflex by returning
 a single, structured map instead.
 
-Read `.agents/skills/devrites-lib/reference/standards/core.md` first — chiefly its vocabulary / existing-conventions
+Read `.agents/skills/devrites-lib/reference/standards/core.md` first: chiefly its vocabulary / existing-conventions
 disciplines, which keep the map in the project's own language. The other rule files load
 on demand.
 
 ## What this skill returns
 
-A **structural map** — terse. One pass should answer:
+A **structural map**: terse. One pass should answer:
 
-- **The area** — what this code is for, in one sentence, using the project's own
+- **The area:** what this code is for, in one sentence, using the project's own
   vocabulary.
-- **Modules in scope** — the related files / packages / slices, with a one-line
+- **Modules in scope:** the related files / packages / slices, with a one-line
   purpose each.
-- **Callers (in)** — who calls into this area from outside. Keep to the highest-signal
-  3–6; collapse the rest.
-- **Calls (out)** — what this area depends on downstream.
-- **Decisions touching it** — ADRs (under `docs/adr/` if present) or notes in
+- **Callers (in):** who calls into this area from outside. Keep to the highest-signal
+  3-6; collapse the rest.
+- **Calls (out):** what this area depends on downstream.
+- **Decisions touching it:** ADRs (under `docs/adr/` if present) or notes in
   `.devrites/work/<slug>/decisions.md` that pre-decide something here.
-- **Smallest sensible change-scope** — where a fix would naturally land, so the next
+- **Smallest sensible change-scope:** where a fix would naturally land, so the next
   step doesn't drift into a project-wide refactor.
 
 ## Prefer a code-intelligence index (if available)
 
-If the project has them — `codebase-memory-mcp` (`get_architecture` / `search_graph`) first,
-cross-checked with `codegraph` (`.codegraph/`) and `graphify` (`graphify-out/`) — use them. For
-codegraph, `codegraph_context` + one `codegraph_explore` return the map in two calls — vastly
+If the project has them, start with `codebase-memory-mcp` (`get_architecture` / `search_graph`),
+then cross-check with `codegraph` (`.codegraph/`) and `graphify` (`graphify-out/`). For
+codegraph, `codegraph_context` + one `codegraph_explore` return the map in two calls: vastly
 cheaper than a file-walk and more accurate for callers/callees. Fall back to standard methods
 (LSP, then `Grep` + `Read`) when no index is available. See `.agents/skills/devrites-lib/reference/standards/tooling.md`.
 
 ## Vocabulary discipline
 
-Use the **project's** domain language — `CONTEXT.md`, glossaries, the active feature's
+Use the **project's** domain language: `CONTEXT.md`, glossaries, the active feature's
 `spec.md` / `decisions.md`. Don't invent fresh names for things the project already
 names. If you notice a fuzzy or overloaded term while mapping, flag it as a FYI at the
 end; don't try to fix it here.
 
 ## When NOT to use
 
-- You already have a clear mental model — zooming out is just tax.
-- The question is a literal text lookup (a string, a comment, an error message) — use
+- You already have a clear mental model: zooming out is just tax.
+- The question is a literal text lookup (a string, a comment, an error message): use
   `Grep`.
-- You need to design or change something — that's `$rite-spec` (new feature) or
+- You need to design or change something. That's `$rite-spec` (new feature) or
   `$rite-define` (plan an approved spec).
-- You want a project-wide architecture audit — use the project's normal architecture
+- You want a project-wide architecture audit: use the project's normal architecture
   review process; this skill is a read-only feature-area map.
 
 ## Output shape

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "devrites",
-  "version": "3.0.5",
-  "description": "DevRites — disciplined senior-engineer workflow skills pack for Claude Code and Codex",
+  "version": "3.0.6",
+  "description": "DevRites: a disciplined senior-engineer workflow pack for Claude Code and Codex",
   "license": "SEE LICENSE IN LICENSE",
   "homepage": "https://github.com/ViktorsBaikers/DevRites#readme",
   "repository": {

package/scripts/build-binaries.sh

@@ -1,6 +1,6 @@
 #!/usr/bin/env bash
 # Cross-compile devrites-engine for every supported target and emit a sha256
-# sidecar per binary — the artifacts attached to the GitHub Release alongside the
+# sidecar per binary: the artifacts attached to the GitHub Release alongside the
 # pack tarball. Pure-Go (CGO_ENABLED=0) so every target builds
 # from one Linux runner with no cross-toolchain. The version is stamped into the
 # binary via -ldflags so `devrites-engine doctor` reports the release it came from.

package/scripts/build-host-artifacts.sh

@@ -1,5 +1,5 @@
 #!/usr/bin/env bash
-# build-host-artifacts.sh — render host-native Claude and Codex artifacts from
+# build-host-artifacts.sh: render host-native Claude and Codex artifacts from
 # the canonical pack.
 #
 # Default output is pack/generated/ so npm pack can ship prebuilt surfaces.

package/scripts/build-release-tarball.sh

@@ -1,5 +1,5 @@
 #!/usr/bin/env bash
-# Build the DevRites release tarball — the artifact attached to the GitHub Release
+# Build the DevRites release tarball: the artifact attached to the GitHub Release
 # by semantic-release. Extracting yields a `devrites-v<version>/` directory with
 # everything an end-user needs (pack/, install.sh, uninstall.sh, scripts/, docs).
 #

package/scripts/check-cross-refs.py

@@ -5,11 +5,11 @@ Catches the maintenance-drift class the skills audit surfaced: a SKILL.md or
 reference file pointing at another file that has moved, been renamed, or never
 existed. Three checks, tuned for near-zero false positives:
 
-  1. Markdown links to local .md targets  — `](path.md)` resolved relative to the
+  1. Markdown links to local .md targets : `](path.md)` resolved relative to the
      containing file; the target must exist.
-  2. Bare backtick .md filenames          — `` `name.md` `` must exist SOMEWHERE in
+  2. Bare backtick .md filenames         : `` `name.md` `` must exist SOMEWHERE in
      the pack (by basename). Catches references to a file that exists nowhere.
-  3. "canonical version: `path`" claims    — the named path (relative to the skills
+  3. "canonical version: `path`" claims   : the named path (relative to the skills
      root) must exist. Catches a file disclaiming itself secondary to a missing one.
 
 Scans pack/.claude/. Exit 1 on any dead reference. Read-only.
@@ -35,7 +35,7 @@ WORKSPACE_ARTIFACTS = {
     "questions.md", "drift.md", "evidence.md", "browser-evidence.md", "touched-files.md",
     "review.md", "seal.md", "ship.md", "brief.md", "design-brief.md", "strategy.md",
     "polish-report.md", "handoff.md", "eng-review.md", "test-plan.md", "references.md",
-    # forge competition record (work/<slug>/forge-report.md) — written when a Forge: yes slice competes candidates
+    # forge competition record (work/<slug>/forge-report.md): written when a Forge: yes slice competes candidates
     "forge-report.md",
     "conventions.md", "coverage.md", "analysis.md", "learnings.md",
     # developer-experience scorecard (work/<slug>/devex.md) + project-level retro ledger (.devrites/retro.md)
@@ -107,11 +107,11 @@ for path in md_files:
         base = os.path.basename(tok)
         if base in WORKSPACE_ARTIFACTS or base in PROJECT_DOCS or base in all_basenames:
             continue
-        if "/" in tok:  # a path inside the pack must actually resolve
+        if "/" in tok:  # a path inside the pack must resolve
             if not os.path.isfile(resolve(here, tok)):
                 errors.append(f"{path}: backtick `{tok}` does not resolve")
             continue
-        errors.append(f"{path}: backtick `{tok}` — no `{base}` exists anywhere in the pack")
+        errors.append(f"{path}: backtick `{tok}`: no `{base}` exists anywhere in the pack")
 
     # 4. "(see X.md)" prose pointers: slash form resolves (install-aware);
     #    bare form must exist in the pack (or be a workspace artifact / project doc).
@@ -122,7 +122,7 @@ for path in md_files:
             if not os.path.isfile(resolved):
                 errors.append(f"{path}: (see {tok}) (resolved {resolved}) does not exist")
         elif base not in WORKSPACE_ARTIFACTS and base not in PROJECT_DOCS and base not in all_basenames:
-            errors.append(f"{path}: (see {tok}) — no `{base}` exists anywhere in the pack")
+            errors.append(f"{path}: (see {tok}): no `{base}` exists anywhere in the pack")
 
     # 3. "canonical version: `path`" must resolve against the skills root
     for claim in CANONICAL_RE.findall(text):
@@ -136,4 +136,4 @@ if errors:
         print("  " + e)
     sys.exit(1)
 
-print(f"check-cross-refs: OK — {len(md_files)} markdown files, no dead references.")
+print(f"check-cross-refs: OK: {len(md_files)} markdown files, no dead references.")

package/scripts/check-invocation-integrity.py

@@ -1,10 +1,10 @@
 #!/usr/bin/env python3
-"""Invocation-integrity gate — every skill/rule an agent or skill NAMES must resolve, both harnesses.
+"""Invocation-integrity gate: every skill/rule an agent or skill NAMES must resolve, both harnesses.
 
 `check-cross-refs.py` validates markdown links. This validates the thing it can't: a **bare
 skill-name invocation** (`$devrites-frontend-craft`, `Skill(devrites-source-driven)`, "invoke the
 `devrites-api-interface` skill") and a **rule read** (`.claude/skills/devrites-lib/reference/standards/security.md`) must each point
-at something that actually exists — on Claude Code AND on the Codex-translated tree — or the agent
+at something that exists (on Claude Code AND on the Codex-translated tree) or the agent
 calls into the void at runtime. Also asserts the Claude→Codex skill/rule mirror is complete.
 
 Self-contained: installs DevRites into a temp dir to get the Codex tree, sweeps, cleans up.
@@ -15,7 +15,7 @@ import os, re, sys, glob, subprocess, tempfile, shutil
 REPO = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 PACK = os.path.join(REPO, "pack/.claude")
 
-# devrites-* tokens that are scripts/hooks/mcp, not skills — legitimate non-skill references.
+# devrites-* tokens that are scripts/hooks/mcp, not skills: legitimate non-skill references.
 NONSKILL = {
     "devrites-mcp", "devrites-lib", "devrites-reviewer-readonly", "devrites-wright-scope",
     "devrites-a1-guard", "devrites-allow", "devrites-redwatch", "devrites-orient",
@@ -103,7 +103,7 @@ def main():
         if problems or mirror:
             print("INVOCATION-INTEGRITY: FAIL")
             return 1
-        print("INVOCATION-INTEGRITY: PASS — every named skill/rule resolves on both harnesses.")
+        print("INVOCATION-INTEGRITY: PASS: every named skill/rule resolves on both harnesses.")
         return 0
     finally:
         shutil.rmtree(target, ignore_errors=True)

package/scripts/check-no-global-writes.sh

@@ -1,5 +1,5 @@
 #!/usr/bin/env bash
-# check-no-global-writes.sh — static guard that the installer/uninstaller never
+# check-no-global-writes.sh: static guard that the installer/uninstaller never
 # writes to the user's global ~/.claude or ~/.codex, and that the global guard is present.
 # Exits non-zero on violation.
 
@@ -40,7 +40,7 @@ done
 #    engine-binary path (the devrites-engine control-plane binary; issue 10). Anything else
 #    writing to /usr/bin, /usr/local/bin, /bin, /sbin, or ~/.local/bin is a global
 #    write the installer must not do. The carve-out is exactly a `devrites-engine` (or
-#    `devrites.exe`) leaf under /usr/local/bin or ~/.local/bin — never /usr/bin.
+#    `devrites.exe`) leaf under /usr/local/bin or ~/.local/bin: never /usr/bin.
 BIN_RE='((\$HOME|~)/\.local/bin|/usr/local/bin|/usr/bin|/bin|/sbin)'
 # Sanctioned: the write's bin path ends in .../devrites-engine (optionally .exe) and sits
 # under /usr/local/bin or ~/.local/bin. The trailing boundary keeps `devrites-lib`

package/scripts/check-reply-contract.sh

@@ -1,5 +1,5 @@
 #!/usr/bin/env bash
-# check-reply-contract.sh — keep user-facing rite-* completion replies normalized.
+# check-reply-contract.sh: keep user-facing rite-* completion replies normalized.
 
 set -u
 

package/scripts/check-rule-uniqueness.sh

@@ -1,9 +1,9 @@
 #!/usr/bin/env bash
-# scripts/check-rule-uniqueness.sh — assert each canonical principle heading
+# scripts/check-rule-uniqueness.sh: assert each canonical principle heading
 # appears in exactly one canonical file under pack/.claude/.
 #
 # Drift catcher: if someone re-duplicates a principle's full treatment, this
-# script fails and names the offenders. Cross-link summaries are fine — they
+# script fails and names the offenders. Cross-link summaries are fine: they
 # don't use the canonical heading.
 #
 # Usage:

package/scripts/codex-generate.sh

@@ -1,5 +1,5 @@
 #!/usr/bin/env bash
-# codex-generate.sh — shared Claude-to-Codex surface generators.
+# codex-generate.sh: shared Claude-to-Codex surface generators.
 #
 # This file is a shell library, not a standalone installer. Callers provide
 # TMP_GEN_DIR and then invoke the gen_* helpers below.
@@ -47,10 +47,10 @@ gen_codex_skill_file() {
       print ""
       print "- Load DevRites engineering standards from `.agents/skills/devrites-lib/reference/standards/`. Read `.agents/skills/devrites-lib/reference/standards/core.md` before workflow work, then load the other `.agents/skills/devrites-lib/reference/standards/*.md` files exactly when this skill asks for them."
       print "- Use the installed `devrites-engine` binary as the canonical runtime helper surface for orientation, gates, and state mutation."
-      print "- When this skill asks for a DevRites specialist or writer agent, **explicitly** spawn the matching Codex custom agent from `.codex/agents/devrites-*.toml` through Codex subagents (`spawn_agent`), then wait for its result and reconcile it as the skill instructs. Do not do the review inline just because the instruction to spawn is embedded here — Codex under-fires embedded spawn/skill instructions (openai/codex #23496), so treat the spawn as required, not optional."
-      print "- The independence of a fresh-context subagent is the point. If Codex genuinely cannot spawn subagents in the current surface, run the documented inline fallback and **label the result an inline fallback, not an independent review** — an inline pass shares the calling context and is weaker evidence."
+      print "- When this skill asks for a DevRites specialist or writer agent, **explicitly** spawn the matching Codex custom agent from `.codex/agents/devrites-*.toml` through Codex subagents (`spawn_agent`), then wait for its result and reconcile it as the skill instructs. Do not do the review inline just because the instruction to spawn is embedded here: Codex under-fires embedded spawn/skill instructions (openai/codex #23496), so treat the spawn as required, not optional."
+      print "- The independence of a fresh-context subagent is the point. If Codex genuinely cannot spawn subagents in the current surface, run the documented inline fallback and **label the result an inline fallback, not an independent review**: an inline pass shares the calling context and is weaker evidence."
       print "- Codex project hooks are installed in `.codex/hooks.json`. Review and trust them with `/hooks` before relying on hook enforcement."
-      print "- When this skill asks a HITL question via `AskUserQuestion`: Codex'\''s equivalent (`request_user_input`) exists only in Plan mode. Outside Plan mode, render the option set as a plain numbered list in chat and **end the turn** so the human answers — NEVER silently pick an option yourself; auto-picking is AFK'\''s contract, gated by the `.devrites/AFK` sentinel."
+      print "- When this skill asks a HITL question via `AskUserQuestion`: Codex'\''s equivalent (`request_user_input`) exists only in Plan mode. Outside Plan mode, render the option set as a plain numbered list in chat and **end the turn** so the human answers: NEVER silently pick an option yourself; auto-picking is AFK'\''s contract, gated by the `.devrites/AFK` sentinel."
       print ""
       fm = 0
       inserted = 1
@@ -293,7 +293,7 @@ gen_codex_hooks_json() {
   }
 }
 EOF
-# NOTE: no SessionEnd entry — Codex does not emit a SessionEnd hook event (its lifecycle events
+# NOTE: no SessionEnd entry: Codex does not emit a SessionEnd hook event (its lifecycle events
 # are SessionStart/SubagentStart/PreToolUse/PermissionRequest/PostToolUse/PreCompact/PostCompact/
 # UserPromptSubmit/SubagentStop/Stop). The `hook event session-end` trace is Claude-only; re-add
 # here if Codex ever ships the event.

package/scripts/devrites-detect.sh

@@ -1,5 +1,5 @@
 #!/usr/bin/env bash
-# scripts/devrites-detect.sh — deterministic anti-slop detector.
+# scripts/devrites-detect.sh: deterministic anti-slop detector.
 #
 # Greps the current git diff (or a passed file list) for ~25 actionable
 # anti-slop patterns from rite-polish/reference/anti-ai-slop.md +
@@ -76,43 +76,43 @@ fi
 # Severity = critical (auth/safety), high (slop tell), low (style nit).
 RULES=(
   # --- UI anti-slop (CSS-ish) ---
-  '#000([^0-9a-f]|$)|#fff([^0-9a-f]|$)|#000000|#ffffff'                            'high'  'Pure #000 / #fff — use near-black / near-white tokens.'
-  '#6366f1|#8b5cf6|#a855f7|#d946ef|#ec4899'                                        'high'  'Purple/blue gradient hex range — likely default AI palette.'
-  'background-clip:[[:space:]]*text|-webkit-background-clip:[[:space:]]*text'      'high'  'Gradient text decoration — likely AI slop.'
-  'backdrop-filter:[[:space:]]*blur'                                               'high'  'Glassmorphism `backdrop-filter: blur(...)` — banned default surface.'
-  'border-left:[[:space:]]*[1-6]px[[:space:]]+solid|border-l-[2-8]'                'low'   'Side-stripe accent border — common templating tell; confirm vs design system.'
-  'cubic-bezier\([^)]*0\.34[[:space:]]*,[[:space:]]*1\.56'                         'high'  'Bounce/elastic easing curve — banned unless design system uses it.'
-  'animation-duration:[[:space:]]*[5-9][0-9]{2}ms[[:space:]]*;[[:space:]]*animation-delay'  'low'   'Long animation durations stacked — verify motion budget.'
-  'z-index:[[:space:]]*[0-9]{4,}'                                                  'high'  'Raw z-index in the 1000s — use semantic z scale.'
-  'font-family:[[:space:]]*"?(DM Sans|Plus Jakarta|Fraunces|Newsreader)'           'high'  'Reflex font choice — use the project type system.'
-  'text-transform:[[:space:]]*uppercase[[:space:]]*;[^}]*letter-spacing'           'low'   'All-CAPS with letter-spacing — verify register; banned for body text.'
+  '#000([^0-9a-f]|$)|#fff([^0-9a-f]|$)|#000000|#ffffff'                            'high'  'Pure #000 / #fff: use near-black / near-white tokens.'
+  '#6366f1|#8b5cf6|#a855f7|#d946ef|#ec4899'                                        'high'  'Purple/blue gradient hex range: likely default AI palette.'
+  'background-clip:[[:space:]]*text|-webkit-background-clip:[[:space:]]*text'      'high'  'Gradient text decoration: likely AI slop.'
+  'backdrop-filter:[[:space:]]*blur'                                               'high'  'Glassmorphism `backdrop-filter: blur(...)`: banned default surface.'
+  'border-left:[[:space:]]*[1-6]px[[:space:]]+solid|border-l-[2-8]'                'low'   'Side-stripe accent border: common templating tell; confirm vs design system.'
+  'cubic-bezier\([^)]*0\.34[[:space:]]*,[[:space:]]*1\.56'                         'high'  'Bounce/elastic easing curve: banned unless design system uses it.'
+  'animation-duration:[[:space:]]*[5-9][0-9]{2}ms[[:space:]]*;[[:space:]]*animation-delay'  'low'   'Long animation durations stacked: verify motion budget.'
+  'z-index:[[:space:]]*[0-9]{4,}'                                                  'high'  'Raw z-index in the 1000s: use semantic z scale.'
+  'font-family:[[:space:]]*"?(DM Sans|Plus Jakarta|Fraunces|Newsreader)'           'high'  'Reflex font choice: use the project type system.'
+  'text-transform:[[:space:]]*uppercase[[:space:]]*;[^}]*letter-spacing'           'low'   'All-CAPS with letter-spacing: verify register; banned for body text.'
   # --- Code anti-slop ---
-  '\bconsole\.log\('                                                               'high'  'console.log left in code — remove before polish.'
-  'if[[:space:]]*\([[:space:]]*[a-zA-Z_][a-zA-Z0-9_]*[[:space:]]*&&[[:space:]]*[a-zA-Z_][a-zA-Z0-9_]*\.length[[:space:]]*>[[:space:]]*0' 'high' 'Over-defensive `x && x.length > 0` — validate at boundaries.'
-  '\bcatch[[:space:]]*\([[:space:]]*\)[[:space:]]*\{[^}]*\}'                       'high'  'Blanket catch with no narrow handling — likely swallowing errors.'
-  '\bcatch[[:space:]]*\([[:space:]]*[A-Za-z][A-Za-z0-9_]*[[:space:]]*\)[[:space:]]*\{[[:space:]]*\}'  'high'  'Empty catch body — silently swallows errors. Catch narrow; recover or rethrow.'
-  '//[[:space:]]*helper function|//[[:space:]]*increment'                          'low'   'Tutorial-style comment — restates code; remove.'
-  '^[[:space:]]*(//|#)[[:space:]]*(Now (I('"'"'ll| will)|we('"'"'ll| will))|Updated|Added|Changed|Modified|As requested|Step [0-9]+:)'  'low'   'Diff-narrating comment — describe the current constraint or delete it.'
-  '\bfunction[[:space:]]+(process|handle|do)(Data|Item|Thing|It)?\b'               'high'  'Generic AI naming (`processData` / `handleItem` / `doIt`) — name for intent.'
-  '\bfunction[[:space:]]+get[A-Z][a-zA-Z]*\([^)]*\)[[:space:]]*\{[[:space:]]*return[[:space:]]+[A-Z][a-zA-Z]+\.find'  'low'   'Useless wrapper around a single ORM call — inline.'
-  '//[[:space:]]*TODO:[[:space:]]*(improve|cleanup|refactor)[[:space:]]+(this|later)'  'low'   '"TODO: improve this later" — without owner/issue, this rots.'
-  '\btemp\s*=|\bresult\s*=|\bdata\s*=[^=]'                                         'low'   'Generic local name (temp/result/data) — name for intent.'
-  '\binterface[[:space:]]+I[A-Z][a-zA-Z]+\b'                                       'low'   '`IFoo` prefix on interfaces — most modern style guides prefer `Foo`.'
-  '/\*\s*eslint-disable\s*\*/'                                                     'high'  'Global eslint-disable — explain or remove.'
+  '\bconsole\.log\('                                                               'high'  'console.log left in code: remove before polish.'
+  'if[[:space:]]*\([[:space:]]*[a-zA-Z_][a-zA-Z0-9_]*[[:space:]]*&&[[:space:]]*[a-zA-Z_][a-zA-Z0-9_]*\.length[[:space:]]*>[[:space:]]*0' 'high' 'Over-defensive `x && x.length > 0`: validate at boundaries.'
+  '\bcatch[[:space:]]*\([[:space:]]*\)[[:space:]]*\{[^}]*\}'                       'high'  'Blanket catch with no narrow handling: likely swallowing errors.'
+  '\bcatch[[:space:]]*\([[:space:]]*[A-Za-z][A-Za-z0-9_]*[[:space:]]*\)[[:space:]]*\{[[:space:]]*\}'  'high'  'Empty catch body: silently swallows errors. Catch narrow; recover or rethrow.'
+  '//[[:space:]]*helper function|//[[:space:]]*increment'                          'low'   'Tutorial-style comment: restates code; remove.'
+  '^[[:space:]]*(//|#)[[:space:]]*(Now (I('"'"'ll| will)|we('"'"'ll| will))|Updated|Added|Changed|Modified|As requested|Step [0-9]+:)'  'low'   'Diff-narrating comment: describe the current constraint or delete it.'
+  '\bfunction[[:space:]]+(process|handle|do)(Data|Item|Thing|It)?\b'               'high'  'Generic AI naming (`processData` / `handleItem` / `doIt`): name for intent.'
+  '\bfunction[[:space:]]+get[A-Z][a-zA-Z]*\([^)]*\)[[:space:]]*\{[[:space:]]*return[[:space:]]+[A-Z][a-zA-Z]+\.find'  'low'   'Useless wrapper around a single ORM call: inline.'
+  '//[[:space:]]*TODO:[[:space:]]*(improve|cleanup|refactor)[[:space:]]+(this|later)'  'low'   '"TODO: improve this later": without owner/issue, this rots.'
+  '\btemp\s*=|\bresult\s*=|\bdata\s*=[^=]'                                         'low'   'Generic local name (temp/result/data): name for intent.'
+  '\binterface[[:space:]]+I[A-Z][a-zA-Z]+\b'                                       'low'   '`IFoo` prefix on interfaces: most modern style guides prefer `Foo`.'
+  '/\*\s*eslint-disable\s*\*/'                                                     'high'  'Global eslint-disable: explain or remove.'
   # --- Security ---
-  '\bnew Function\('                                                               'critical' '`new Function(...)` — dynamic code construction.'
-  '\beval\s*\('                                                                    'critical' '`eval(...)` — avoid.'
-  '\bdangerouslySetInnerHTML\b'                                                    'high'  'dangerouslySetInnerHTML — confirm XSS-safe.'
-  'process\.env\.[A-Z_]+(.{0,40}(?:console|res\.send|res\.json))'                  'high'  'env var possibly echoed to response/log — re-check for secret leakage.'
+  '\bnew Function\('                                                               'critical' '`new Function(...)`: dynamic code construction.'
+  '\beval\s*\('                                                                    'critical' '`eval(...)`: avoid.'
+  '\bdangerouslySetInnerHTML\b'                                                    'high'  'dangerouslySetInnerHTML: confirm XSS-safe.'
+  'process\.env\.[A-Z_]+(.{0,40}(?:console|res\.send|res\.json))'                  'high'  'env var possibly echoed to response/log: re-check for secret leakage.'
   # --- Performance pitfalls ---
-  '\.forEach\([^)]+\)\s*\.\s*forEach\('                                            'low'   'Nested `forEach` — accidental quadratic loop?'
+  '\.forEach\([^)]+\)\s*\.\s*forEach\('                                            'low'   'Nested `forEach`: accidental quadratic loop?'
 )
 
 # Markdown-only rules. Run against SCAN_MD, not SCAN.
-# Em-dash overuse: 2+ em-dashes (U+2014) on the same line — the per-line proxy
+# Em-dash overuse: 2+ em-dashes (U+2014) on the same line: the per-line proxy
 # for the "multiple em-dashes per paragraph" rule in anti-ai-slop.md.
 MD_RULES=(
-  $'.*\xe2\x80\x94.*\xe2\x80\x94'  'high'  'Em-dash overuse on one line (2+ em-dashes) — a common AI tell; prefer comma, period, or parenthetical.'
+  $'.*\xe2\x80\x94.*\xe2\x80\x94'  'high'  'Em-dash overuse on one line (2+ em-dashes): a common AI tell; prefer comma, period, or parenthetical.'
 )
 
 FOUND=0
@@ -133,7 +133,7 @@ if [[ ${#SCAN[@]} -gt 0 ]]; then
     while IFS= read -r line; do
       [[ -z "$line" ]] && continue
       FOUND=$((FOUND + 1))
-      printf '[%s] %s — %s\n' "$sev" "$line" "$msg"
+      printf '[%s] %s: %s\n' "$sev" "$line" "$msg"
     done <<<"$HITS"
   done
 fi
@@ -154,7 +154,7 @@ if [[ ${#SCAN_MD[@]} -gt 0 ]]; then
     while IFS= read -r line; do
       [[ -z "$line" ]] && continue
       FOUND=$((FOUND + 1))
-      printf '[%s] %s — %s\n' "$sev" "$line" "$msg"
+      printf '[%s] %s: %s\n' "$sev" "$line" "$msg"
     done <<<"$HITS"
   done
 fi
@@ -163,7 +163,7 @@ TOTAL=$(( ${#SCAN[@]} + ${#SCAN_MD[@]} ))
 
 echo
 if [[ $FOUND -eq 0 ]]; then
-  echo "devrites-detect: clean — scanned $TOTAL files (${#SCAN[@]} code, ${#SCAN_MD[@]} md)."
+  echo "devrites-detect: clean: scanned $TOTAL files (${#SCAN[@]} code, ${#SCAN_MD[@]} md)."
   exit 0
 fi
 

package/scripts/eval-runner.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 """DevRites trigger-eval runner.
 
-Wraps the Anthropic SDK to actually execute a trigger eval against a Claude
+Wraps the Anthropic SDK to execute a trigger eval against a Claude
 model. Off by default in CI (schema-only validation lives in
 `scripts/run-evals.sh`); this runner is invoked explicitly when
 `CLAUDE_API_KEY` is set.
@@ -26,10 +26,10 @@ What it does per query
 1. Loads each skill's `name` + `description` from `pack/.claude/skills/*/SKILL.md`.
 2. Sends the query as a `user` message to Claude with a system prompt
    instructing it to pick exactly one skill name from the candidate list, or
-   "none" if no skill applies. We do NOT ask the model to invoke a tool — we
+   "none" if no skill applies. We do NOT ask the model to invoke a tool: we
    ask it to predict which skill the harness would have triggered. That keeps
    the eval cheap (no agentic loop, one round-trip per query) and faithful to
-   what skill discovery actually decides.
+   what skill discovery decides.
 3. Compares the predicted skill name to the expected verdict:
    - `should_trigger`  → predicted name must equal the eval's `skill` field.
    - `should_not_trigger` → predicted name must NOT equal the eval's `skill`.
@@ -45,7 +45,7 @@ Usage
     CLAUDE_API_KEY=sk-... ./scripts/eval-runner.py --min-accuracy 0.9 evals/*.json
 
 The runner is intentionally thin (~200 lines). It is not the eval-viewer in
-Anthropic's `skill-creator` — it just provides the missing "did the model
+Anthropic's `skill-creator`: it just provides the missing "did the model
 pick the right skill" signal so we can answer that question without firing
 up the full Claude Code harness.
 """
@@ -67,7 +67,7 @@ SYSTEM_TEMPLATE = """\
 You are the DevRites skill router. Your job is to predict which DevRites
 skill (if any) should fire for a given user message.
 
-Each candidate skill is described below. Read the descriptions carefully —
+Each candidate skill is described below. Read the descriptions carefully:
 DevRites' triggers are encoded in them.
 
 For the user message at the end, respond with exactly one line of the form:
@@ -113,7 +113,7 @@ def load_skills() -> list[Skill]:
         if not skill_md.is_file():
             continue
         body = skill_md.read_text(encoding="utf-8")
-        # Lightweight YAML frontmatter parse — we only need name + description.
+        # Lightweight YAML frontmatter parse: we only need name + description.
         m = re.match(r"^---\n(.*?)\n---", body, re.DOTALL)
         if not m:
             continue
@@ -129,7 +129,7 @@ def load_skills() -> list[Skill]:
 def render_system(skills: list[Skill]) -> str:
     lines = []
     for s in skills:
-        lines.append(f"- **{s.name}** — {s.description}")
+        lines.append(f"- **{s.name}**: {s.description}")
     return SYSTEM_TEMPLATE.format(skills="\n".join(lines))