@quickback-dev/cli 0.66.5 → 0.66.6

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.
Files changed (77) hide show
  1. package/README.md +0 -1
  2. package/dist/index.js +1 -1
  3. package/dist/skill/SKILL.md +0 -2
  4. package/dist/skill/docs/api/openapi.md +1 -1
  5. package/dist/skill/docs/configure/auth.md +4 -5
  6. package/dist/skill/docs/configure/domains.md +1 -1
  7. package/dist/skill/docs/configure/index.md +4 -4
  8. package/dist/skill/docs/configure/security.md +3 -3
  9. package/dist/skill/docs/define/access.md +23 -3
  10. package/dist/skill/docs/define/actions/defining.md +2 -2
  11. package/dist/skill/docs/define/actions/record-and-standalone.md +1 -2
  12. package/dist/skill/docs/define/areas.md +14 -15
  13. package/dist/skill/docs/define/changesets.md +5 -0
  14. package/dist/skill/docs/define/diagnostics.md +2 -2
  15. package/dist/skill/docs/define/fga.md +1 -1
  16. package/dist/skill/docs/define/firewall.md +5 -5
  17. package/dist/skill/docs/define/guards.md +2 -1
  18. package/dist/skill/docs/define/masking.md +2 -2
  19. package/dist/skill/docs/define/relationships.md +8 -10
  20. package/dist/skill/docs/define/scopes.md +1 -1
  21. package/dist/skill/docs/index.md +0 -6
  22. package/dist/skill/docs/platform/auth/jwt-optimization.md +2 -2
  23. package/dist/skill/docs/platform/realtime/durable-objects.md +1 -1
  24. package/dist/skill/docs/platform/realtime/partyserver-rooms.md +4 -4
  25. package/dist/skill/docs/platform/webhooks/inbound.md +1 -1
  26. package/dist/skill/docs/tooling/claude-skill.md +0 -7
  27. package/dist/skill/docs/tooling/cloud-compiler/endpoints.md +1 -1
  28. package/dist/skill/docs/tooling/output.md +7 -6
  29. package/dist/skill/docs/ui/account/features/admin.md +1 -1
  30. package/dist/skill/docs/ui/account/features/organizations.md +1 -1
  31. package/dist/skill/docs/ui/admin/index.md +1 -1
  32. package/dist/skill/docs/ui/admin/owned-records.md +81 -0
  33. package/dist/skill/docs/ui/admin/schema-format.md +16 -0
  34. package/package.json +1 -1
  35. package/src/skill/SKILL.md +0 -2
  36. package/src/skill/docs/api/openapi.md +1 -1
  37. package/src/skill/docs/configure/auth.md +4 -5
  38. package/src/skill/docs/configure/domains.md +1 -1
  39. package/src/skill/docs/configure/index.md +4 -4
  40. package/src/skill/docs/configure/security.md +3 -3
  41. package/src/skill/docs/define/access.md +23 -3
  42. package/src/skill/docs/define/actions/defining.md +2 -2
  43. package/src/skill/docs/define/actions/record-and-standalone.md +1 -2
  44. package/src/skill/docs/define/areas.md +14 -15
  45. package/src/skill/docs/define/changesets.md +5 -0
  46. package/src/skill/docs/define/diagnostics.md +2 -2
  47. package/src/skill/docs/define/fga.md +1 -1
  48. package/src/skill/docs/define/firewall.md +5 -5
  49. package/src/skill/docs/define/guards.md +2 -1
  50. package/src/skill/docs/define/masking.md +2 -2
  51. package/src/skill/docs/define/relationships.md +8 -10
  52. package/src/skill/docs/define/scopes.md +1 -1
  53. package/src/skill/docs/index.md +0 -6
  54. package/src/skill/docs/platform/auth/jwt-optimization.md +2 -2
  55. package/src/skill/docs/platform/realtime/durable-objects.md +1 -1
  56. package/src/skill/docs/platform/realtime/partyserver-rooms.md +4 -4
  57. package/src/skill/docs/platform/webhooks/inbound.md +1 -1
  58. package/src/skill/docs/tooling/claude-skill.md +0 -7
  59. package/src/skill/docs/tooling/cloud-compiler/endpoints.md +1 -1
  60. package/src/skill/docs/tooling/output.md +7 -6
  61. package/src/skill/docs/ui/account/features/admin.md +1 -1
  62. package/src/skill/docs/ui/account/features/organizations.md +1 -1
  63. package/src/skill/docs/ui/admin/index.md +1 -1
  64. package/src/skill/docs/ui/admin/owned-records.md +81 -0
  65. package/src/skill/docs/ui/admin/schema-format.md +16 -0
  66. package/dist/skill/docs/changelog/index.md +0 -945
  67. package/dist/skill/docs/changelog/v0.1x.md +0 -1202
  68. package/dist/skill/docs/changelog/v0.2x.md +0 -589
  69. package/dist/skill/docs/changelog/v0.3x.md +0 -1152
  70. package/dist/skill/docs/changelog/v0.4x.md +0 -1108
  71. package/dist/skill/docs/changelog/v0.5x.md +0 -885
  72. package/src/skill/docs/changelog/index.md +0 -945
  73. package/src/skill/docs/changelog/v0.1x.md +0 -1202
  74. package/src/skill/docs/changelog/v0.2x.md +0 -589
  75. package/src/skill/docs/changelog/v0.3x.md +0 -1152
  76. package/src/skill/docs/changelog/v0.4x.md +0 -1108
  77. package/src/skill/docs/changelog/v0.5x.md +0 -885
package/README.md CHANGED
@@ -118,4 +118,3 @@ MIT © Kardoe Incorporated
118
118
  - [Quickback compiler docs](https://docs.quickback.dev)
119
119
  - [GitHub repository](https://github.com/paulstenhouse/quickback-compiler)
120
120
  - [Issue tracker](https://github.com/paulstenhouse/quickback-compiler/issues)
121
- - [Changelog](https://docs.quickback.dev/compiler/changelog)
package/dist/index.js CHANGED
@@ -10,7 +10,7 @@
10
10
  */
11
11
  import pc from "picocolors";
12
12
  // Version injected at build time by scripts/inject-version.ts
13
- const CLI_VERSION = "0.66.5";
13
+ const CLI_VERSION = "0.66.6";
14
14
  function getPackageVersion() {
15
15
  return CLI_VERSION;
16
16
  }
@@ -443,6 +443,4 @@ The path layout mirrors the public docs site (https://docs.quickback.dev). Open
443
443
  - `claude-skill` / `claude-code` — this skill
444
444
  - `cloud-compiler/{index,authentication,endpoints,local-compiler,troubleshooting}` — the hosted build service
445
445
 
446
- **Other** — `docs/changelog/index.md`
447
-
448
446
  For online navigation: <https://docs.quickback.dev>.
@@ -232,7 +232,7 @@ Or enable the in-project browser via the [`docs` config option](#interactive-api
232
232
 
233
233
  Alongside the spec, every compile writes `reports/operations.manifest.json` — a
234
234
  stable, machine-readable index of the full route surface (resource, action,
235
- view, and auth routes; since v0.47 also the realtime, storage, and FGA
235
+ view, auth, realtime, storage, and FGA routes, plus
236
236
  surfaces):
237
237
 
238
238
  ```json
@@ -55,7 +55,7 @@ The short version, before the knob-by-knob detail:
55
55
  - **Default: 180-second TTL, auto-refreshed.** Every authenticated call returns a fresh token via the `set-auth-token` header, so the low TTL is invisible to logged-in users. The worst-case window for a stolen or pre-revocation token is 3 minutes — and you can shrink it to 30 seconds with one config line.
56
56
  - **Need revocation faster than the TTL?** Set [`revocationCheck: 'kv'`](#revocationcheck-none--kv--token-revocation-strategy). Revoked tokens are rejected within seconds; revocation stamps are written automatically on sign-out and session revocation (per session), on password change with `revokeOtherSessions`, account deletion, user ban, org-member removal, role change, and changes to scope-conferring relationship rows. Long-lived tokens are therefore safe to pair with it: a remote "sign out everywhere" kills a lost device's bearer.
57
57
 
58
- Those two knobs — `expiresIn` and `revocationCheck` — cover the vast majority of deployments. The honest residual for the strictest threat models: the KV check revokes per-principal (not per-session) and propagates cross-PoP in ~60s. It **fails closed** — if KV can't be read, the bearer isn't trusted and a bearer-only caller gets a retryable `503`, so enabling it ties bearer availability to KV. If you need per-session kill semantics, set `expiresIn: 30` for a hard 30-second ceiling — and [tell us about your use case](https://github.com/kardoe/quickback/issues); an opaque-session mode is under consideration.
58
+ Those two knobs — `expiresIn` and `revocationCheck` — cover the vast majority of deployments. The honest residual for the strictest threat models: the KV check revokes per-principal (not per-session) and propagates cross-PoP in ~60s. It **fails closed** — if KV can't be read, the bearer isn't trusted and a bearer-only caller gets a retryable `503`, so enabling it ties bearer availability to KV. If you need per-session kill semantics, set `expiresIn: 30` for a hard 30-second ceiling.
59
59
 
60
60
  ### Behavior recap
61
61
 
@@ -130,7 +130,7 @@ Default `'none'`.
130
130
  | `'none'` | No revocation check. Stolen / pre-revocation tokens valid until TTL expires (scope claims: until the carry-forward cap). Use a low `expiresIn` to bound the replay window. | 0 |
131
131
  | `'kv'` | Fast-path checks per-principal revocation timestamps in the project's existing `KV` binding before trusting a verified token. Fails closed: an unreadable store means the bearer is not trusted. | 2+ KV reads (~5-15ms, edge-cached) |
132
132
 
133
- `'kv'` (v0.45+, **Cloudflare runtime only**) keeps the fast-path stateless against the auth DB (still 0 D1 queries). There is nothing extra to provision — entries ride the `KV` binding every project already has, prefixed `qbrev:` and self-expiring after the maximum token lifetime, so the store never accumulates.
133
+ `'kv'` (**Cloudflare runtime only**) keeps the fast-path stateless against the auth DB (still 0 D1 queries). There is nothing extra to provision — entries ride the `KV` binding every project already has, prefixed `qbrev:` and self-expiring after the maximum token lifetime, so the store never accumulates.
134
134
 
135
135
  A token is rejected when its mint timestamp (`iat`/`iatMs`, or the scope claim's proof time) is **at or before** the stored timestamp for any principal key it rides on:
136
136
 
@@ -147,8 +147,7 @@ Semantics worth knowing:
147
147
 
148
148
  - **Revoke-on-touch for scopes:** any update or delete of a conferring row stamps the revocation, without evaluating whether the change actually ended the relationship. A false positive just forces a cheap re-prove (`/scope/v1/enter` or a minting namespace route).
149
149
  - **Scope proof time is per kind:** a token can carry several scope kinds proven at different instants, so each entry carries its own proof stamp (`sctMs`, `m<ms>`) and is compared against its own marker. Proving one kind never restamps the others — re-entering `event` cannot bring a revoked `location` claim back to life. Entries minted before the release that added the stamp carry none, and fall back to the token-level `sct`/`sctMs` until they expire (bounded by the carry-forward cap plus one TTL).
150
- - **No same-second gap:** markers are stored in epoch milliseconds (written `m<ms>` — an explicit format version, not a magnitude the reader guesses at) and tokens carry millisecond companions (`iatMs` / `sctMs`) alongside the standards-compatible seconds claims, so a token minted a few milliseconds before the revocation is rejected — while one legitimately re-minted *after* it in that same second is still accepted. Tokens and markers written by v0.60.4 or earlier are handled **conservatively**: with no sub-second information on one side, the whole marker second is revoked. That over-revocation is bounded — the last old token expires within its TTL, and the last old marker within the entry lifetime. Like any rejected token it costs the caller a `401`, not a silent recovery: the bearer falls through unauthenticated (its cookie having been stripped, per the next bullet), and the client re-authenticates for a fresh token.
151
- - **Do not straddle the v0.61.0 upgrade.** A worker from **before** v0.61.0 reads `m<ms>` markers as `NaN` and treats them as *no revocation at all* — not as a corrupt entry. So rolling back across this release, or pointing two differently-versioned deploy targets at one shared KV namespace, means every revocation written by the new version is **silently unenforced** by the old one until the entry expires (`ENTRY_TTL_SECONDS`, ~19 minutes at default settings). Upgrade all targets sharing a namespace together; if you must roll back, treat outstanding bearer tokens as unrevoked for that window — lower `expiresIn`, or purge the `qbrev:` keys and re-stamp from the old version.
150
+ - **No same-second gap:** markers are stored in epoch milliseconds (written `m<ms>` — an explicit format marker, not a magnitude the reader guesses at) and tokens carry millisecond companions (`iatMs` / `sctMs`) alongside the standards-compatible seconds claims, so a token minted a few milliseconds before the revocation is rejected — while one legitimately re-minted *after* it in that same second is still accepted.
152
151
  - **Fails closed:** a missing binding, a failed read, or a corrupt entry means the token's revocation state is *unknown* — and an unknown bearer is not authenticated. The caller gets `503 AUTH_REVOCATION_UNAVAILABLE` (RFC 9457, retryable). Recovery is a **separate cookie-only request**: a request carrying `Authorization` has its cookie stripped before session auth, so within that one request there is nothing left to fall back to. Reads are never silently skipped, because the outage that hides a revocation is exactly the moment this check exists for.
153
152
  - **Availability coupling:** the flip side of the above. Bearer-token auth now depends on KV being readable. Alert on `AUTH_REVOCATION_UNAVAILABLE`; a sustained rate means the binding is missing on some deployment target, or KV is degraded. Every named environment must declare `bindings.kv` — the compiler rejects a target that doesn't.
154
153
  - **Eventual consistency:** KV propagates cross-PoP in ~60s. "Immediate" means seconds at the writing location, bounded-seconds globally — still far inside the 15-minute carry-forward cap that bounds revocation without this check.
@@ -195,7 +194,7 @@ The `?redirect=…` query param is preserved unchanged so the central sign-in su
195
194
 
196
195
  Signing in is the *only* thing the CMS gate redirects for. A signed-in caller who isn't admitted is served the CMS shell and shown a screen naming what they'd need — see [CMS access](/ui/admin#access-gate). There is no wrong-role redirect, and no `auth.profileUrl`.
197
196
 
198
- > **Removed in v0.62:** `auth.profileUrl`. Its only consumer was the CMS wrong-role redirect. The config schema is strict, so a leftover `profileUrl` key now fails the compile — delete it.
197
+ > `auth.profileUrl` is not supported. The config schema is strict, so a `profileUrl` key fails the compile — delete it.
199
198
 
200
199
 
201
200
  ### Resolution precedence
@@ -280,7 +280,7 @@ EXTRA_APP_HOSTS = "app-preview.example.com,pr-42.example.com"
280
280
  wrangler secret put EXTRA_APP_HOSTS --env preview
281
281
  ```
282
282
 
283
- Mirrors the existing `BETTER_AUTH_URL` auto-trust behavior, where the deployed worker auto-trusts its own configured base origin without needing an explicit `trustedOrigins` entry. With `EXTRA_APP_HOSTS`, one env var extends all three allowlists — apps, CORS, and Better Auth's `trustedOrigins`. (Earlier v0.64 output stopped at the CORS layer: preflight passed, then Better Auth's own check rejected the origin. On older output, recompile — or set `BETTER_AUTH_URL` on the deploy target, since Better Auth always trusts its own base URL.)
283
+ Mirrors the existing `BETTER_AUTH_URL` auto-trust behavior, where the deployed worker auto-trusts its own configured base origin without needing an explicit `trustedOrigins` entry. With `EXTRA_APP_HOSTS`, one env var extends all three allowlists — apps, CORS, and Better Auth's `trustedOrigins`.
284
284
 
285
285
  > **Multi-app deploys**
286
286
  >
@@ -145,7 +145,7 @@ a compile error; delete that line and keep `routes`. See
145
145
  [API Contract](/platform/api-contract#contractversion-is-retired) for what
146
146
  changes if you were pinned to `version: "v1"`.
147
147
 
148
- Action input-schema fidelity is a compile gate since v0.63.0. The CLI must
148
+ Action input-schema fidelity is a compile gate. The CLI must
149
149
  successfully harvest `z.toJSONSchema(action.input)` for every authored action;
150
150
  partial or missing harvests fail with the exact action file keys, bundle or
151
151
  evaluation phase, and underlying local error. Dependencies are resolved from
@@ -433,7 +433,7 @@ action layer.
433
433
 
434
434
  Quickback manages four **audit columns** (`createdAt`, `modifiedAt`, `createdBy`, `modifiedBy`) on every feature table, plus two **soft-delete columns** (`deletedAt`, `deletedBy`) on tables whose resource uses soft delete (the default delete mode). The audit columns drive record-provenance queries and the CMS audit panel; the soft-delete pair drives soft-delete filtering.
435
435
 
436
- You declare them in the table's `columns` object with `...q.audit()` and `...q.softDelete()`. The compiler validates every managed column against its canonical shape and fails the compile when one is missing, diverges, or is present on a hard-delete table — run `quickback migrate visible-columns` to adopt the declarations on an existing project. Set `compiler.features.visibleAuditColumns: false` to downgrade those checks to warnings and keep the pre-v0.57.0 silent-injection behavior.
436
+ You declare them in the table's `columns` object with `...q.audit()` and `...q.softDelete()`. The compiler validates every managed column against its canonical shape and fails the compile when one is missing, diverges, or is present on a hard-delete table — run `quickback migrate visible-columns` to adopt the declarations on an existing project. Set `compiler.features.visibleAuditColumns: false` to downgrade those checks to warnings and silently inject the columns.
437
437
 
438
438
  To opt out of the managed columns entirely, project-wide:
439
439
 
@@ -485,7 +485,7 @@ export default defineConfig({
485
485
 
486
486
  If `signature.required: true` and no key is available, compilation fails with a clear error message (or warning in `failMode: "warn"`).
487
487
 
488
- ## Hostname architecture (v0.33+)
488
+ ## Hostname architecture
489
489
 
490
490
  A modern Quickback project runs on **one Cloudflare Worker** that serves multiple URL surfaces. By default, everything lives under a single unified backend hostname (auto-inferred as `quickback.{baseDomain}` from your other configured subdomains, or overridable via `config.domain`). For larger deployments, you can opt-in to **dedicated hostnames** for specific surfaces — they're additive aliases, never replacements.
491
491
 
@@ -515,7 +515,7 @@ For branding or operational separation, you can opt specific surfaces onto their
515
515
  cms: { domain: 'cms.example.com' },
516
516
  account: { domain: 'account.example.com' },
517
517
 
518
- // v0.33+ — backend surfaces on their own hosts:
518
+ // Backend surfaces on their own hosts:
519
519
  auth: { domain: 'auth.example.com' },
520
520
  api: { domain: 'api.example.com' },
521
521
 
@@ -118,7 +118,7 @@ The compiler applies these so config stays minimal:
118
118
  - **`X-Frame-Options` derives from `csp.frameAncestors`** (`['@none']` → `DENY`, contains `@self` → `SAMEORIGIN`, otherwise omitted — XFO can't represent arbitrary origins so CSP becomes the truth there). Override with `security.frameOptions`.
119
119
  - **`corp: 'same-origin'` is auto-emitted when `coep` is set.** `require-corp` without CORP breaks the ASSETS binding fetches on Cloudflare; the compiler couples them so users don't have to remember.
120
120
 
121
- ### CSP auto-config (0.10.15+)
121
+ ### CSP auto-config
122
122
 
123
123
  When you opt into a `csp` object, the compiler additively widens directives based on which Quickback features your project has enabled. **Additive** means anything you already wrote stays; we only push values that aren't already present.
124
124
 
@@ -140,7 +140,7 @@ font-src 'self' https://fonts.gstatic.com
140
140
  connect-src 'self' @trustedOrigins https://api.your-app.com
141
141
  ```
142
142
 
143
- Each directive is scoped to exactly what it needs. `defaultSrc` is *not* widened with style-only or connect-only hosts — that would let those hosts also reach `script-src` / `img-src` / `media-src` via fall-back, which is wider than the bundled SPA actually needs. (This rule changed in 0.10.17 — earlier versions incorrectly pushed additions into defaultSrc.)
143
+ Each directive is scoped to exactly what it needs. `defaultSrc` is *not* widened with style-only or connect-only hosts — that would let those hosts also reach `script-src` / `img-src` / `media-src` via fall-back, which is wider than the bundled SPA actually needs.
144
144
 
145
145
  **Visibility:** every applied addition prints to stderr at compile time — one line per addition with the directive, value, and reason. No silent magic. Loop's compile output now looks like:
146
146
 
@@ -155,7 +155,7 @@ Each directive is scoped to exactly what it needs. `defaultSrc` is *not* widened
155
155
 
156
156
  #### Why the SPA-baked origin shows up capitalised in CSP
157
157
 
158
- Projects with `domain: "..."` + bundled CMS / Account SPAs will see `connect-src 'self' https://Api.example.com …` in their served CSP — note the capital `A`. That's deliberate (added in 0.10.19) and works around a `wrangler dev` quirk, not a typo on our end.
158
+ Projects with `domain: "..."` + bundled CMS / Account SPAs will see `connect-src 'self' https://Api.example.com …` in their served CSP — note the capital `A`. That's deliberate and works around a `wrangler dev` quirk, not a typo on our end.
159
159
 
160
160
  `wrangler dev` rewrites the configured `custom_domain` host string in **every** response header value (Location, Content-Security-Policy, X-…) to the local hostname, with a case-sensitive substring matcher. Without the case-shift, the served CSP becomes `connect-src 'self' https://localhost:8787 …` and the bundled SPA's hardcoded `VITE_QUICKBACK_API_URL` fetches get blocked because the prod URL no longer appears in the allowlist.
161
161
 
@@ -100,7 +100,7 @@ any lowercase role name you reference in an access tree (`"recruiter"`,
100
100
  `"hiring-manager"`, `"support"`, …) is a custom org-membership role, matched
101
101
  against `ctx.roles` like the built-ins.
102
102
 
103
- As of v0.46, custom roles declared in access trees are **auto-registered
103
+ Custom roles declared in access trees are **auto-registered
104
104
  with the Better Auth organization plugin**, so members can actually be
105
105
  provisioned into them: `invite-member` and `update-member-role` accept them
106
106
  (previously they rejected custom roles with `400 ROLE_NOT_FOUND`).
@@ -132,6 +132,26 @@ interface Access {
132
132
  }
133
133
  ```
134
134
 
135
+ ### An empty list is a compile error
136
+
137
+ `roles: []`, `userRole: []`, and `principals: []` are refused at compile
138
+ time, on every access node — CRUD operations, `read`, views, actions, and
139
+ the arms of `or` / `and`. The runtime evaluator treats an empty list as
140
+ "no gate on this axis", so an arm that names nobody would admit **every
141
+ authenticated caller**. Masking uses the same literal the other way round
142
+ (`show: { roles: [] }` shows nobody), which is exactly why the compiler will
143
+ not guess here. Say what you mean:
144
+
145
+ {/* doc-compile: skip — config fragments, not a compilable project */}
146
+ ```typescript
147
+ create: { access: { userRole: ['sysadmin'] } } // closed to everyone but platform admins
148
+ // or omit `create` entirely — the route is not emitted
149
+ read: { access: { roles: ['AUTHENTICATED'] } } // every signed-in caller really is the audience
150
+ ```
151
+
152
+ Every empty arm in the project is listed in one compile error, so a project
153
+ fixes them all in one pass.
154
+
135
155
  ## `userRole` vs `roles`
136
156
 
137
157
  `roles` and `userRole` check **two different fields** and are often confused — get this wrong and you'll grant access too broadly.
@@ -200,7 +220,7 @@ access: { roles: ["PUBLIC"] }
200
220
  - `PUBLIC` skips authentication and role checks — anyone can call the endpoint
201
221
  - **`PUBLIC` is the explicit, uppercase opt-in for unauthenticated access.** It works on every access node — reads, top-level writes (`create`/`update`/`delete`/`upsert`), views, and actions. The compiler trusts the marker; row-level safety comes from the firewall, masking, and (for token-gated endpoints like RSVP confirm, magic-link redemption, List-Unsubscribe-Post, webhook ingest) your handler's own validation. Anonymous write endpoints typically pair `roles: ["PUBLIC"]` with `firewall: { exception: true }` on the resource so the row predicate doesn't reject every unauthenticated caller.
202
222
  - **Firewall still applies** — data is scoped by the firewall (organization, owner, team)
203
- - **Tree semantics:** `PUBLIC` admits anonymous callers from anywhere in the access tree, not just a flat `roles` list. An `or:` arm containing `PUBLIC` admits anonymous callers past the authentication gate; an `and:` group admits them only when *every* arm does. The access, org, and firewall layers still evaluate the full tree afterwards. (Before v0.46 the emitted auth gate only honored `PUBLIC` in a flat `roles` list — a nested `PUBLIC` arm incorrectly answered `401`.)
223
+ - **Tree semantics:** `PUBLIC` admits anonymous callers from anywhere in the access tree, not just a flat `roles` list. An `or:` arm containing `PUBLIC` admits anonymous callers past the authentication gate; an `and:` group admits them only when *every* arm does. The access, org, and firewall layers still evaluate the full tree afterwards.
204
224
  - Every `PUBLIC` action invocation is **mandatory audit logged** to the security audit table (IP address, input, result, timing)
205
225
  - The wildcard `"*"` is **not supported** — using it throws a compile-time error
206
226
  - Tables with no isolation columns and any `PUBLIC` route do not require `firewall: [{ exception: true }]`
@@ -330,7 +350,7 @@ Use one of these instead:
330
350
 
331
351
  #### `SYSADMIN` — cross-tenant DB-admin tier
332
352
 
333
- Use `roles: ["SYSADMIN"]` for endpoints that should only be reachable by platform sysadmins — the cross-tenant DB-admin tier introduced in v0.15.0. Compiles to `ctx.userRole === "sysadmin"`. **Requires `cms: { sysadmin: true }` on the project config**; without that flag, using `SYSADMIN` fails the compile so the role can't sit as dead code.
353
+ Use `roles: ["SYSADMIN"]` for endpoints that should only be reachable by platform sysadmins — the cross-tenant DB-admin tier. Compiles to `ctx.userRole === "sysadmin"`. **Requires `cms: { sysadmin: true }` on the project config**; without that flag, using `SYSADMIN` fails the compile so the role can't sit as dead code.
334
354
 
335
355
  ```typescript
336
356
  // Cross-tenant audit dashboard, sysadmin-only
@@ -115,8 +115,8 @@ Zod 4's `z.toJSONSchema`. It does not call `execute`; the hosted compiler never
115
115
  evaluates project code. The output contract feeds OpenAPI and
116
116
  `Api.<feature>.actions.<action>.Output` in `quickback.gen.ts`.
117
117
 
118
- **Every action input must produce a schema.** Since
119
- v0.63.0 a failed harvest fails the compile, naming the exact action, whether
118
+ **Every action input must produce a schema.** A failed harvest fails the
119
+ compile, naming the exact action, whether
120
120
  bundling or evaluation failed, and the underlying local error. Output remains
121
121
  optional.
122
122
 
@@ -17,7 +17,7 @@ POST /applications/:id/schedule-interview
17
17
  1. **Authentication** - User token is validated
18
18
  2. **Input Validation** - Request body validated against Zod schema
19
19
  3. **Role Gate** - The access roles are checked *before* the record is
20
- fetched (v0.46+), so a wrong-role caller gets a uniform
20
+ fetched, so a wrong-role caller gets a uniform
21
21
  `403 ACCESS_ROLE_REQUIRED` whether or not the record exists — the
22
22
  response can't be used to probe for record ids
23
23
  4. **Record Loading** - The record is fetched by ID **through the firewall**
@@ -420,4 +420,3 @@ imply "this writes data" — it's the right method whenever the request body is
420
420
  load-bearing. Use the response shape (`responseType: "json"`, or just the data
421
421
  your handler returns) to convey read semantics; HTTP method choice is about
422
422
  input transport, not action intent.
423
-
@@ -305,7 +305,7 @@ doubt: if only one folder's features reference it, it belongs in that folder's a
305
305
 
306
306
  When several standalone actions share a path prefix and the same relationship gate — `/event/:eventId/announce`, `/event/:eventId/cancel`, `/event/:eventId/respond` — declaring the gate per-action means writing the same `roles: ['attendee']` over and over AND paying the relationship-resolver cost N times per request (one subquery + one row-load, multiplied by the number of actions).
307
307
 
308
- Namespace mounts hoist the resolver into one Hono middleware that runs once per request. Every action under the prefix skips the inline gate; `ctx.<exposeAs>` is hydrated by the middleware before the action's `execute()` runs. With the per-namespace helper (v0.27+, [below](#narrow-ctxexposeas-typing-v027)), that field is also statically typed as the loaded row.
308
+ Namespace mounts hoist the resolver into one Hono middleware that runs once per request. Every action under the prefix skips the inline gate; `ctx.<exposeAs>` is hydrated by the middleware before the action's `execute()` runs. With the [per-namespace helper](#narrow-ctxexposeas-typing), that field is also statically typed as the loaded row.
309
309
 
310
310
  ### When to use roles vs. relationships
311
311
 
@@ -317,13 +317,13 @@ A user is authorized for a row when **any** of the following is true:
317
317
 
318
318
  Use **relationships** when membership is specific to *this row* — external collaborators who aren't in your org, attendees, anyone connected to the resource by virtue of data rather than payroll. Use **roles / team** when staff at your org/team should see everything in their scope by default.
319
319
 
320
- The two compose at the action level (`access.or` accepts both relationship-roles and tenant roles) and — as of v0.25 — at the namespace level too: `namespace.via` accepts an array of lanes, first match wins.
320
+ The two compose at the action level (`access.or` accepts both relationship-roles and tenant roles) and at the namespace level: `namespace.via` accepts an array of lanes, first match wins.
321
321
 
322
322
  > **Rule of thumb.** If the answer is *"yes if you work here,"* reach for org/team roles. If the answer is *"yes if you have something to do with **this** row,"* reach for a per-row relationship. Reach for both together — at the namespace or via `access.or` — when an action should let either side in.
323
323
 
324
324
  Two declaration sites depending on whether the participating actions live in one feature or many:
325
325
 
326
- ### Single feature: `defineTable({ namespace })` (v0.23+)
326
+ ### Single feature: `defineTable({ namespace })`
327
327
 
328
328
  Use when every action under the prefix lives in the same feature directory. The namespace travels with the table that owns it — no addition to `quickback.config.ts` needed.
329
329
 
@@ -366,9 +366,9 @@ export default defineAction({
366
366
 
367
367
  The compiler emits one `app.use(...)` on the events feature's Hono app. Every standalone action in `features/events/actions/` whose path starts with `/event/:eventId/` joins the namespace automatically.
368
368
 
369
- Multi-lane via (v0.25+) and bulk-grant lanes (v0.26+) work identically on the table-level declaration — `via:` accepts the same shapes documented below for the project-level form.
369
+ Multi-lane and bulk-grant lanes work identically on the table-level declaration — `via:` accepts the same shapes documented below for the project-level form.
370
370
 
371
- ### Cross-feature: `authz.namespaces` (v0.24+)
371
+ ### Cross-feature: `authz.namespaces`
372
372
 
373
373
  Use when the namespace spans multiple features. The declaration sits next to your other authz config in `quickback.config.ts`.
374
374
 
@@ -422,7 +422,7 @@ export default defineAction({
422
422
 
423
423
  The compiler emits one synthetic `src/routes/__namespace-event-scope.routes.ts` that imports each contributing feature's `actions` map under a distinct alias, mounts the resolver middleware once, and dispatches to the matching action. Worker-root mount order ensures the middleware fires before any per-feature route handler under the prefix.
424
424
 
425
- ### Multi-lane via (v0.25+) — first match wins
425
+ ### Multi-lane via — first match wins
426
426
 
427
427
  When more than one relationship can authorize a caller for the namespaced resource, pass `via:` as an array. The middleware tries each lane in declaration order and stops at the first match — `ctx.<exposeAs>` is hydrated from whichever lane fired.
428
428
 
@@ -458,9 +458,9 @@ authz: {
458
458
  }
459
459
  ```
460
460
 
461
- Hydration is **per matched lane** (v0.37+): each lane fills `ctx.<its own exposeAs>` from its own `loads:` table, so lanes may load different tables into different keys. Both lanes above name `events` / `event`, so `ctx.event` is populated whichever fires. A lane declaring neither `loads:` nor `exposeAs:` is authorization-only — it opens the gate and hydrates nothing.
461
+ Hydration is **per matched lane**: each lane fills `ctx.<its own exposeAs>` from its own `loads:` table, so lanes may load different tables into different keys. Both lanes above name `events` / `event`, so `ctx.event` is populated whichever fires. A lane declaring neither `loads:` nor `exposeAs:` is authorization-only — it opens the gate and hydrates nothing.
462
462
 
463
- ### Bulk-grant lanes (v0.26+)
463
+ ### Bulk-grant lanes
464
464
 
465
465
  Relationship lanes cover "specifically connected to this row." Bulk-grant lanes cover "anyone with this role / team membership in the row's tenant" — the case where an org admin or a team member should see every event in their scope without writing per-event relationship rows.
466
466
 
@@ -483,7 +483,7 @@ Each bulk-grant lane:
483
483
  - **`{ roles: [...] }`** — passes when the caller's BA org-role for the loaded resource's tenant includes any listed role. Short-circuits the relationship loop entirely; the role check is in-memory and cheap. Use for the "company admin sees everything" case.
484
484
  - **`{ team: true }`** — passes when `ctx.activeTeamId` is set. The loads-table firewall does the actual `events.teamId === ctx.activeTeamId` check during hydration. Use when the resource is team-scoped and you want active-team membership to grant access.
485
485
 
486
- Both bulk-grant lanes piggyback on the namespace's relationship lane(s) for hydration — they don't fetch a separate per-row subject; they just gate access and let the loads-table firewall handle the tenant scoping. **At least one relationship lane is required** alongside bulk-grant lanes in v0.26 (to anchor `loads:` + `exposeAs:`).
486
+ Both bulk-grant lanes piggyback on the namespace's relationship lane(s) for hydration — they don't fetch a separate per-row subject; they just gate access and let the loads-table firewall handle the tenant scoping. When a relationship lane anchors hydration, declare `loads:` + `exposeAs:` on that lane.
487
487
 
488
488
  When a request misses every lane, the 403 deny payload lists every authorization path the caller could have taken — relationship names, role names, and `"team-membership"` for `{ team: true }` lanes — so a developer debugging an unexpected 403 sees the full surface they need to satisfy.
489
489
 
@@ -497,7 +497,7 @@ Actions opt in by matching `path:` against a declared namespace prefix. There's
497
497
 
498
498
  Both forms share the same shape checks:
499
499
 
500
- - **`via:` accepts one of three shapes.** A single relationship name (v0.23+), an array of relationship names (v0.25+, first match wins), or — as of v0.26 — an array mixing relationship names with bulk-grant guards (`{ roles: [...] }`, `{ team: true }`). **A bare string is always a relationship name.** There is no string form for a role: `{ roles: [...] }` is a bulk-grant lane against Better Auth org roles, even when the name you put in it is a composed role defined `via` a relationship.
500
+ - **`via:` accepts one of three shapes.** A single relationship name, an array of relationship names (first match wins), or an array mixing relationship names with bulk-grant guards (`{ roles: [...] }`, `{ team: true }`). **A bare string is always a relationship name.** There is no string form for a role: `{ roles: [...] }` is a bulk-grant lane against Better Auth org roles, even when the name you put in it is a composed role defined `via` a relationship.
501
501
  - **Each relationship lane declares `loads` + `exposeAs`, or neither.** Both → the lane hydrates `ctx.<exposeAs>` from `loads` when it matches. Neither → an authorization-only lane. Exactly one is a compile error.
502
502
  - **Bulk-grant lanes inherit `loads:` + `exposeAs:` from sibling relationship lanes.** Pure bulk-grant namespaces (no relationship lane at all) are also accepted — set `loads:` + `exposeAs:` directly on the namespace declaration so hydration has somewhere to anchor.
503
503
  - **`prefix:` must contain at least one `:param` segment** (e.g. `/event/:eventId`). The middleware reads the relationship key from `c.req.param()` because no request body has been parsed at middleware time.
@@ -516,11 +516,11 @@ Inside a namespace action's `execute()`:
516
516
  - Tenant context (`ctx.activeOrgId` / `ctx.activeTeamId`) is stamped from the matched relationship row when its `from` table carries those columns. Downstream scoped queries work without further setup.
517
517
  - The action's own `access:` block (additional roles, userRole, record predicates, OTHER relationship arms) still runs after the namespace gate — namespace handles entry, per-action gates further.
518
518
 
519
- ### Narrow `ctx.<exposeAs>` typing (v0.27+)
519
+ ### Narrow `ctx.<exposeAs>` typing
520
520
 
521
- The namespace middleware hydrates `ctx.<exposeAs>` at runtime — that's been true since v0.23. The static type of that field, though, is `any` via the `AppContext` index signature: useful for not blocking access, but a missed opportunity for autocomplete and typo-catching.
521
+ The namespace middleware hydrates `ctx.<exposeAs>` at runtime. The static type of that field is `any` via the `AppContext` index signature: useful for not blocking access, but a missed opportunity for autocomplete and typo-catching.
522
522
 
523
- v0.27 emits one **per-namespace `defineAction` helper** alongside the standard `<feature>/.quickback/define-action.ts`. Importing from the narrowed helper retypes `ctx.<exposeAs>` inside `execute({ ctx })` to the loaded row's `$inferSelect` shape. Nothing else changes — runtime behavior, action discovery, and the `path:` namespace match are identical.
523
+ The compiler emits one **per-namespace `defineAction` helper** alongside the standard `<feature>/.quickback/define-action.ts`. Importing from the narrowed helper retypes `ctx.<exposeAs>` inside `execute({ ctx })` to the loaded row's `$inferSelect` shape. Nothing else changes — runtime behavior, action discovery, and the `path:` namespace match are identical.
524
524
 
525
525
  ```ts title="features/feeds/actions/feedMoments.ts"
526
526
  // One-line flip per namespace-scoped action:
@@ -549,7 +549,7 @@ The compiler emits one helper per `(feature, namespace)` pair where the feature
549
549
 
550
550
  #### Opt-in, not breaking
551
551
 
552
- Actions that keep their existing `import { defineAction } from '../.quickback/define-action'` continue to compile unchanged — `ctx.<exposeAs>` is `any` via the `AppContext` index signature, which is exactly the v0.23–v0.26 behavior. You flip imports per action at your own pace.
552
+ Actions that import `defineAction` from `../.quickback/define-action` compile with `ctx.<exposeAs>` as `any` via the `AppContext` index signature. Use the per-namespace helper for the narrowed type.
553
553
 
554
554
  #### Compile-time advisory
555
555
 
@@ -578,4 +578,3 @@ src/features/interviews/.quickback/
578
578
  #### Why per-namespace and not per-action
579
579
 
580
580
  A `defineAction` helper that's narrow to *one* exposed key (the namespace's `exposeAs`) is enough — every action under a namespace sees the same hydrated key. The alternative (per-action emit) duplicates the same prelude N times for the same shape; per-namespace keeps emission proportional to the number of distinct hydration shapes in the project.
581
-
@@ -64,6 +64,11 @@ the operations manifest.
64
64
  update: { access: { roles: ["member", "admin", "owner"] }, routes: false },
65
65
  ```
66
66
 
67
+ The registry records these ops in `routesDisabled`, so the
68
+ [CMS](/ui/admin/owned-records) suppresses raw-CRUD buttons for them while
69
+ still rendering the relation inline on the root's record page — its writes
70
+ are changeset ops, which `routes: false` deliberately leaves open.
71
+
67
72
  ## The wire contract
68
73
 
69
74
  One blessed path — **RFC 5789 media-type dispatch** on the routes you already
@@ -26,7 +26,7 @@ The first line stays greppable (`[quickback:<channel>] CODE: message`); the
26
26
  source location, fix hint, and — for promoted errors — the exact override
27
27
  recipe render on continuation lines.
28
28
 
29
- ## Errors vs warnings (v0.47+)
29
+ ## Errors vs warnings
30
30
 
31
31
  The checks split into two tiers:
32
32
 
@@ -114,7 +114,7 @@ authz: {
114
114
  {
115
115
  check: 'AUTHZ_CAPABILITY_CEILING',
116
116
  target: 'sessions/scope:event:organizer',
117
- reason: 'organizer grant profile lands with the v2 rollout (TICKET-123)',
117
+ reason: 'temporary exception while the organizer grant profile is implemented',
118
118
  },
119
119
  ],
120
120
  },
@@ -307,7 +307,7 @@ stored tuple whose subject shape isn't allowed for the relation is inert. So a
307
307
  `user:*` tuple on a relation declared `[user]` (no wildcard) can never grant
308
308
  access, even if it was written out-of-band — the model is the ceiling.
309
309
 
310
- ## Limitations (v0.35)
310
+ ## Limitations
311
311
 
312
312
  - `access: { fga }` covers **per-record** operations (`GET /:id`, `update`,
313
313
  `delete`, record-based actions) and **collection filtering** on
@@ -100,9 +100,9 @@ type FirewallPredicate =
100
100
  | { field: string; isNull: true } // column IS NULL
101
101
  | { field: string; in: string[] | string; optional?: boolean } // column IN (...)
102
102
  | { field: string; via: string } // column IN (relationship subquery)
103
- | { field?: string; permission: string } // named authz.permissions reference (v0.35+)
104
- | { all: FirewallPredicate[] } // AND group (v0.35+)
105
- | { any: FirewallPredicate[] } // OR group (v0.35+)
103
+ | { field?: string; permission: string } // named authz.permissions reference
104
+ | { all: FirewallPredicate[] } // AND group
105
+ | { any: FirewallPredicate[] } // OR group
106
106
  | { exception: true }; // opt out of tenant scoping
107
107
 
108
108
  type FirewallConfig = ReadonlyArray<FirewallPredicate>;
@@ -113,7 +113,7 @@ firewallErrorMode?: 'reveal' | 'hide';
113
113
 
114
114
  The top-level array is AND'd together. `firewallErrorMode` lives at the resource top level — it's a presentation-layer choice (403 vs 404), not a query-shape one.
115
115
 
116
- ## Boolean groups: `all` / `any` (v0.35+)
116
+ ## Boolean groups: `all` / `any`
117
117
 
118
118
  A flat array can only express AND. To grant access through **either** of two independent paths — "an org member **OR** an event attendee" — nest an `any` group. The top-level array stays sugar for `all`, so every existing firewall is byte-identical.
119
119
 
@@ -389,7 +389,7 @@ So `firewall: [{ exception: true }]` on the default configuration returns `isNul
389
389
 
390
390
  ## Sysadmin Escape Hatch
391
391
 
392
- When `cms: { sysadmin: true }` is set on the project (introduced in v0.15.0), every `buildFirewallConditions()` emits a runtime branch that drops tenant scopes for `ctx.userRole === 'sysadmin'` callers while keeping the soft-delete predicate:
392
+ When `cms: { sysadmin: true }` is set on the project, every `buildFirewallConditions()` emits a runtime branch that drops tenant scopes for `ctx.userRole === 'sysadmin'` callers while keeping the soft-delete predicate:
393
393
 
394
394
  ```typescript
395
395
  // Generated for an org-scoped table when cms.sysadmin: true
@@ -76,7 +76,8 @@ guards: {
76
76
  > to the table for `db.update(<table>).set({ … })` / `db.insert(<table>).values({ … })`
77
77
  > and refuses an `immutable` column in any update, or a `protected` column
78
78
  > written by an action that is not in its list — with an error naming the action,
79
- > the column, and the fix. The two tiers are therefore distinct: `immutable` is
79
+ > the column, and the fix. Every offending action in the project is listed in
80
+ > that one error, so a single compile surfaces all of them. The two tiers are therefore distinct: `immutable` is
80
81
  > written once at create and never again; if a column needs an action to change
81
82
  > it, it is `protected`, mapped to that action.
82
83
  >
@@ -50,7 +50,7 @@ Whether `customers.email` is PII to hide or the display label of a contact list
50
50
 
51
51
  So the compiler asks. A false positive costs you one line (`cc: false`) — which is also a durable, reviewable record that somebody looked.
52
52
 
53
- It also removes a bug class outright. When detection can never quietly supply a rule, the set of masked columns is the set you declared, so no two surfaces can disagree about it. That exact disagreement was a real defect: through v0.58.3 the response path applied declared-only masking while realtime and live views applied detected-plus-declared, so an auto-detected secret was redacted over WebSocket and returned in **plaintext** on `GET`.
53
+ It also removes a bug class outright. When detection can never quietly supply a rule, the set of masked columns is the set you declared, so no two surfaces can disagree about it.
54
54
 
55
55
  ### What a decision controls
56
56
 
@@ -215,7 +215,7 @@ masking: {
215
215
  }
216
216
  ```
217
217
 
218
- `show: { roles: [] }` has the same effect, but it doesn't say the same thing. An empty role list reads as an oversight, and it is also what authors fall back to when the audience they wanted is **inexpressible** — masking gates on org and scoped roles, so a table gated `access: { userRole: ['sysadmin'] }` cannot name its own audience in a mask. Two different intentions, one spelling, and no way for the next reader to tell them apart.
218
+ `show: { roles: [] }` has the same effect, but it doesn't say the same thing. An empty role list reads as an oversight, and it is also what authors fall back to when the audience they wanted is **inexpressible** — masking gates on org and scoped roles, so a table gated `access: { userRole: ['sysadmin'] }` cannot name its own audience in a mask. Two different intentions, one spelling, and no way for the next reader to tell them apart. Note the asymmetry with access, where an empty list would mean *everyone* — which is why [access refuses it at compile time](/define/access#an-empty-list-is-a-compile-error).
219
219
 
220
220
  Write `'never'` when nobody is the intent. Any other string is a compile error rather than a guess — `show: 'admin'` is a plausible typo, and masking-for-everyone is the wrong thing to assume from it.
221
221
 
@@ -118,7 +118,7 @@ the compiler rejects a relationship-bearing `masking.show` rule and points to a
118
118
  relationship-gated view instead. It never drops the relationship arm or turns
119
119
  it into a broader role grant.
120
120
 
121
- ### Standalone actions: surface keys and `loads` / `exposeAs` (v0.19+)
121
+ ### Standalone actions: surface keys and `loads` / `exposeAs`
122
122
 
123
123
  When a standalone action's `access.roles` references a relationship-role, the compiler emits a runtime gate between input validation and `execute()`. The gate reads the relationship's `resource.column` value from the action surface — either a path parameter (`:eventId`) or an input field (`input.eventId`). If neither is present, the current compiler stops with a clear build error; an action can never ship with a relationship rule it has no key to evaluate.
124
124
 
@@ -132,7 +132,7 @@ authz: {
132
132
  subject: { column: 'userId', equals: 'ctx.userId' },
133
133
  resource: { column: 'eventId' },
134
134
  where: { status: 'confirmed' },
135
- loads: 'events', // ← v0.19+: load matching event row
135
+ loads: 'events', // load matching event row
136
136
  exposeAs: 'event', // ← onto ctx.event for the action body
137
137
  },
138
138
  },
@@ -142,7 +142,7 @@ authz: {
142
142
  }
143
143
  ```
144
144
 
145
- The resolver (v0.20):
145
+ The resolver:
146
146
 
147
147
  1. **Cheap in-memory role check** first — `ctx.roles?.includes('admin')` and friends. If the caller has a qualifying role, this short-circuits and no DB queries fire for the access gate.
148
148
  2. **Firewall-gated relationship match** if the role check failed — `SELECT { organizationId, teamId } FROM guests WHERE userId=ctx.userId AND eventId=input.eventId AND status='confirmed'` with the `guests` table's full firewall AND-applied. If neither role nor relationship qualifies → 403.
@@ -152,24 +152,22 @@ The resolver (v0.20):
152
152
 
153
153
  **No firewall bypass anywhere.** Both queries run with their respective tables' full firewalls. The relationship row's existence is the proof that the caller is authorized; its tenant columns are the source of the stamped claim.
154
154
 
155
- ### Conditional tenant preconditions (v0.20+)
155
+ ### Conditional tenant preconditions
156
156
 
157
- Standalone actions in features with org/team-scoped tables previously emitted unconditional preconditions:
157
+ Standalone actions in features with org/team-scoped tables emit tenant preconditions only when a relationship cannot supply the tenant:
158
158
 
159
159
  ```ts
160
160
  if (!ctx.activeOrgId) return c.json(AccessErrors.noOrg(), 403);
161
161
  if (!ctx.activeTeamId) return c.json(AccessErrors.noTeam(), 403);
162
162
  ```
163
163
 
164
- These ran **before** the relationship resolver, which 403'd attendees whose sessions don't carry an active tenant claim (the canonical event-attendee shape).
165
-
166
- v0.20 makes these preconditions **conditional on relationship supply**. If the access tree contains a relationship arm whose `from` table carries the matching tenant column, the precondition is skipped — the resolver runs first and stamps the value. Pseudo-role short-circuits (PUBLIC / AUTHENTICATED / USER / SYSADMIN) and `firewall: { exception: true }` still bypass the preconditions as before. `ADMIN` is now rejected at compile time.
164
+ If the access tree contains a relationship arm whose `from` table carries the matching tenant column, the precondition is skipped — the resolver runs first and stamps the value. Pseudo-role short-circuits (PUBLIC / AUTHENTICATED / USER / SYSADMIN) and `firewall: { exception: true }` also bypass the preconditions. `ADMIN` is rejected at compile time.
167
165
 
168
166
  Typical event-attendee setup (`guests` table carries `organizationId`): the org precondition is dropped on every action whose access references `attendee` or `eventStaff`. Org admins still get `ctx.activeOrgId` from their session as today; attendees get it from the relationship row, mid-request. Both paths converge on a valid tenant claim before `execute()` runs.
169
167
 
170
168
  ### Requirement on the relationship's `from` table
171
169
 
172
- For the v0.20 tenant-stamping to work, the relationship's `from` table must carry tenant columns:
170
+ For tenant-stamping to work, the relationship's `from` table must carry tenant columns:
173
171
 
174
172
  - For org stamping: an `organizationId` column (or whatever spelling the project's auth-provider defaults pick).
175
173
  - For team stamping: a `teamId` column.
@@ -213,7 +211,7 @@ export default defineAction({
213
211
 
214
212
  `loads` and `exposeAs` are co-required (set both or neither). `exposeAs` must be a valid JS identifier and must not collide with built-in `ctx` fields (`userId`, `activeOrgId`, `activeTeamId`, `db`, etc.).
215
213
 
216
- ### Action firewall escape hatch (v0.19+)
214
+ ### Action firewall escape hatch
217
215
 
218
216
  `defineAction({ firewall: { exception: true } })` opts the action out of the inherited session-key preconditions (`ctx.activeOrgId` / `ctx.activeTeamId` requirements) AND the relationship resolver. The handler receives `unsafeDb` for raw Drizzle access. Use for cross-tenant or admin routes whose surface intentionally doesn't name a session/relationship key:
219
217
 
@@ -672,7 +672,7 @@ The scope claim is trusted between proofs — it is never re-checked against the
672
672
 
673
673
  So removing a guest/staff row takes effect within **cap + one TTL** for a client that keeps polling, and within one TTL once it goes idle. Two ways to tighten that:
674
674
 
675
- - **`auth.jwt.revocationCheck: 'kv'`** (v0.45+, Cloudflare runtime) — sub-TTL revocation: updating or deleting a conferring relationship row stamps a revocation timestamp in KV, and the bearer fast-path rejects a carried scope claim whose proof is at or before it — no same-second gap. Each kind is compared against **its own** proof time (`sctMs` on the entry; older tokens without one fall back to the token-level `sct`/`sctMs`), so a mint of one kind can't carry a revoked kind past its marker. Neither the refresh nor a re-mint ever substitutes its own clock for a proof it didn't take. The check fails closed — a scope claim whose revocation state can't be read is not trusted either. See [auth config](/configure/auth).
675
+ - **`auth.jwt.revocationCheck: 'kv'`** (Cloudflare runtime) — sub-TTL revocation: updating or deleting a conferring relationship row stamps a revocation timestamp in KV, and the bearer fast-path rejects a carried scope claim whose proof is at or before it — no same-second gap. Each kind is compared against **its own** proof time (`sctMs` on the entry; tokens without one fall back to the token-level `sct`/`sctMs`), so a mint of one kind can't carry a revoked kind past its marker. Neither the refresh nor a re-mint ever substitutes its own clock for a proof it didn't take. The check fails closed — a scope claim whose revocation state can't be read is not trusted either. See [auth config](/configure/auth).
676
676
  - **`via` arms** — for tables where any caching is too wide, gate with a `via` arm instead of a bare `ctx.scope.<kind>` arm: `via` emits a live subquery against the relationship table on every request, immune to a stale claim entirely.
677
677
 
678
678
  ## Reference
@@ -76,9 +76,3 @@ The runtime services around your compiled code. Everything you need for a comple
76
76
  - **[Account UI](/ui/account)** — Pre-built auth and profile pages
77
77
 
78
78
  Everything runs on **your own Cloudflare account** — no Quickback infrastructure between you and your data.
79
-
80
- ---
81
-
82
- ### Changelog
83
-
84
- [**Changelog**](/changelog) — what's new in Quickback. Release notes, new features, and improvements.
@@ -110,7 +110,7 @@ The fast-path is stateless by default: it verifies the signature and **trusts th
110
110
 
111
111
  **Lower the TTL.** `auth.jwt.expiresIn: 30` gives a hard 30-second ceiling. Browser SPAs auto-refresh via `set-auth-token` on every authed call, so a low TTL is invisible to logged-in users; only stale server-to-server bearer tokens feel it.
112
112
 
113
- **Or enable sub-TTL revocation.** Set `revocationCheck: 'kv'` (v0.45+, Cloudflare runtime only):
113
+ **Or enable sub-TTL revocation.** Set `revocationCheck: 'kv'` (Cloudflare runtime only):
114
114
 
115
115
  ```ts title="quickback.config.ts"
116
116
  auth: {
@@ -120,7 +120,7 @@ auth: {
120
120
 
121
121
  The fast-path then consults per-principal revocation timestamps in your project's existing `KV` binding (keys prefixed `qbrev:`, self-expiring) *before* trusting a token. A token is rejected when its mint timestamp (`iat`/`iatMs`, or the proof time of the scope kind being checked) is **at or before** the stored timestamp. Cost: 2+ edge-cached KV reads per bearer verify.
122
122
 
123
- There is no same-second gap: markers are epoch milliseconds (written `m<ms>`, v0.61.0+), tokens carry millisecond companions, and the comparison denies a tie — so a token minted a few milliseconds *before* the revocation is rejected, while one legitimately re-minted *after* it in that same second is still accepted. Pre-v0.61.0 workers don't understand that marker format and read it as "not revoked" — don't roll back across the upgrade or share one KV namespace between versions; see [`revocationCheck`](/configure/auth#revocationcheck-none--kv--token-revocation-strategy).
123
+ There is no same-second gap: markers are epoch milliseconds (written `m<ms>`), tokens carry millisecond companions, and the comparison denies a tie — so a token minted a few milliseconds *before* the revocation is rejected, while one legitimately re-minted *after* it in that same second is still accepted.
124
124
 
125
125
  Revocation stamps are written **automatically** on:
126
126
 
@@ -367,7 +367,7 @@ await realtime.update('applications', newRecord, oldRecord, ctx.activeOrgId!, {
367
367
  });
368
368
  ```
369
369
 
370
- **Fail-closed (v0.48+):** an omitted or empty `targetRoles` delivers to
370
+ **Fail-closed:** an omitted or empty `targetRoles` delivers to
371
371
  **nobody** — there is no implicit "everyone" fallback. To reach every
372
372
  authenticated subscriber in the scope, say so explicitly with
373
373
  `targetRoles: ['AUTHENTICATED']`. UPPERCASE pseudo-roles (`PUBLIC`,