diff --git a/.2119/verdicts/REQ-003.4.2.json b/.2119/verdicts/REQ-003.4.2.json
index d1ad49c..3d6cf2d 100644
--- a/.2119/verdicts/REQ-003.4.2.json
+++ b/.2119/verdicts/REQ-003.4.2.json
@@ -1,8 +1,8 @@
 {
-  "reviewId": "REQ-003.4.2--0ba9b9c05c15",
+  "reviewId": "REQ-003.4.2--2558ff342f6b",
   "requirementId": "REQ-003.4.2",
-  "hash": "0ba9b9c05c15",
+  "hash": "2558ff342f6b",
   "verdict": "pass",
-  "summary": "Risks section states the self-pass residual risk plainly and names the committed-verdict, hash-invalidation, and CI re-run mitigations explicitly",
-  "timestamp": "2026-08-08T07:28:03.328Z"
+  "summary": "Risks section states the self-pass residual risk plainly and names the committed-verdict, hash-invalidation, and CI re-run mitigations",
+  "timestamp": "2026-08-08T08:53:49.315Z"
 }
diff --git a/.2119/verdicts/REQ-005.2.6.json b/.2119/verdicts/REQ-005.2.6.json
index 01854dd..07de5d4 100644
--- a/.2119/verdicts/REQ-005.2.6.json
+++ b/.2119/verdicts/REQ-005.2.6.json
@@ -1,8 +1,8 @@
 {
-  "reviewId": "REQ-005.2.6--1c79c4f2f0d0",
+  "reviewId": "REQ-005.2.6--403de8d5f77f",
   "requirementId": "REQ-005.2.6",
-  "hash": "1c79c4f2f0d0",
+  "hash": "403de8d5f77f",
   "verdict": "pass",
-  "summary": "Verify-tag docs state arbitrary-shell execution and package.json-script trust equivalence in the section documenting [verify]",
-  "timestamp": "2026-08-08T07:28:03.467Z"
+  "summary": "Verify-tag docs state arbitrary-shell execution and package.json-script trust equivalence beside the [verify] semantics",
+  "timestamp": "2026-08-08T08:53:49.450Z"
 }
diff --git a/.2119/verdicts/REQ-008.1.2.json b/.2119/verdicts/REQ-008.1.2.json
index 2e45411..55f18a6 100644
--- a/.2119/verdicts/REQ-008.1.2.json
+++ b/.2119/verdicts/REQ-008.1.2.json
@@ -1,8 +1,8 @@
 {
-  "reviewId": "REQ-008.1.2--1fcd1a57103b",
+  "reviewId": "REQ-008.1.2--664c5abc793a",
   "requirementId": "REQ-008.1.2",
-  "hash": "1fcd1a57103b",
+  "hash": "664c5abc793a",
   "verdict": "pass",
-  "summary": "Boundaries paragraph names all three exclusions with rationales and the design.md link, in the intro before the adoption instructions",
-  "timestamp": "2026-08-08T07:28:09.788Z"
+  "summary": "Header block states all three exclusions bolded with the design.md link before any install command; numbered trio restates them before the adoption section",
+  "timestamp": "2026-08-08T08:53:49.587Z"
 }
diff --git a/.2119/verdicts/REQ-008.2.2.json b/.2119/verdicts/REQ-008.2.2.json
index 1874a93..175555b 100644
--- a/.2119/verdicts/REQ-008.2.2.json
+++ b/.2119/verdicts/REQ-008.2.2.json
@@ -1,8 +1,8 @@
 {
-  "reviewId": "REQ-008.2.2--456734b19776",
+  "reviewId": "REQ-008.2.2--5308de176d8b",
   "requirementId": "REQ-008.2.2",
-  "hash": "456734b19776",
+  "hash": "5308de176d8b",
   "verdict": "pass",
   "summary": "README's Diversify-then-audit bullet and scaling.md's cross-provider audit section both advise periodic sweeps plus targeted audits of challenging or high-consequence requirements",
-  "timestamp": "2026-08-08T07:57:56.365Z"
+  "timestamp": "2026-08-08T08:53:49.719Z"
 }
diff --git a/README.md b/README.md
index 948e20b..8273478 100644
--- a/README.md
+++ b/README.md
@@ -1,23 +1,22 @@
 # 2119
 
-**Spec-driven test enforcement for coding agents.** Named for
-[RFC 2119](https://www.rfc-editor.org/rfc/rfc2119) · [unsupervised.com/2119](https://unsupervised.com/2119)
-
-2119 makes the planning → building → testing loop hard to cheat:
-
-1. **Plans become requirements.** Features start as specs in `specs/` — RFC 2119
-   documents where every requirement is a numbered, individually addressable
-   statement with exactly one normative keyword. `2119 lint` enforces the format.
-2. **Requirements become tests.** Every MUST-level requirement needs at least one
-   test annotated with its ID (`// 2119: REQ-001.2.3` — a comment, so it works in
-   any language). `2119 cover` fails on any gap, in either direction.
-3. **Tests get judged.** `2119 review` generates one instruction file per
-   requirement asking a *fresh-context* reviewer — one that did not write the
-   code — a single question: **would these tests fail if this requirement were
-   violated?** Verdicts are recorded with `2119 pass` / `2119 fail`.
-4. **One gate.** `2119 check` = lint + coverage + verdict freshness (a verdict
-   counts only while its hash still matches the content it vouched for). Exit
-   code 0 or it isn't done. Agent hooks, git, and CI all call the same command.
+**Spec-driven test enforcement for coding agents — a gate your agent can't quietly game.** Named for [RFC 2119](https://www.rfc-editor.org/rfc/rfc2119).
+
+- **Plans become numbered requirements** — specs where every MUST is individually addressable, and `2119 lint` enforces the format.
+- **Every requirement gets a test that could actually fail** — `2119 cover` fails on missing tests; a *fresh-context* reviewer judges each one: *would these tests fail if this requirement were violated?*
+- **"Done" becomes one exit code** — `2119 check` gates agent hooks, pre-commit, and CI identically. Verdicts are committed, hash-bound artifacts: edit the code or the requirement, and the approval visibly dies.
+
+2119 is **not** a test runner, **not** a CI replacement, and **not** a security boundary ([why](docs/design.md)).
+
+```bash
+npx rfc2119 init
+```
+
+[unsupervised.com/2119](https://unsupervised.com/2119) · [design decisions](docs/design.md) · [scaling guide](docs/scaling.md) · [live adoption example](https://github.com/tylerwillis/panopticon)
+
+═══════════════════════════════════════════════════════════════════
+
+## 2119 makes the planning → building → testing loop hard to cheat
 
 Here is the gate catching an untested requirement and a stale review:
 
@@ -38,15 +37,18 @@ see the full gate running in another codebase — specs, committed verdicts,
 and the CI check on every PR —
 browse [panopticon](https://github.com/tylerwillis/panopticon).
 
-Three things 2119 is deliberately **not**: a test runner (`check` never
-executes your suite — compose them: `npm test && npx rfc2119 check`), a CI
-replacement (it's one exit code your CI calls), and a security boundary (a
-deliberate cheater is made *conspicuous*, not impossible — see
-[Risks](#risks)). These boundaries are
-[enforced as reviewed requirements](specs/REQ-008-honest-boundaries.md); the
-reasoning lives in [docs/design.md](docs/design.md), and
-[docs/scaling.md](docs/scaling.md) covers hardening for larger or more formal
-projects.
+Three things 2119 is deliberately **not** — each
+[enforced as a reviewed requirement](specs/REQ-008-honest-boundaries.md):
+
+1. **Not a test runner** — `check` never executes your suite; compose them:
+   `npm test && npx rfc2119 check`.
+2. **Not a CI replacement** — it's one exit code your CI calls.
+3. **Not a security boundary** — a deliberate cheater is made *conspicuous*,
+   not impossible (see [Risks](#risks)).
+
+The reasoning lives in [docs/design.md](docs/design.md);
+[docs/scaling.md](docs/scaling.md) covers hardening for larger or more
+formal projects.
 
 ## Use it in your repo