@@ -22,7 +22,7 @@ Because enrichments run on Sim's hosted keys by default, **every provider tool y
 
 ## Architecture (what you're plugging into)
 
-- **`enrichments/types.ts`** — `EnrichmentConfig { id, name, description, icon, inputs, outputs, providers }` and `EnrichmentProvider { id, label, toolId, buildParams, projectFailure, mapOutput }` — `toolProvider` fills `projectFailure` with the standard HTTP projection; override it only for a provider whose failure shape is nonstandard (see `enrichments/provider-failures/`). Providers are **plain data** (no `@/tools` import) so the catalog stays client-safe.
+- **`enrichments/types.ts`** — `EnrichmentConfig { id, name, description, icon, inputs, outputs, providers }` and `EnrichmentProvider { id, label, toolId, buildParams, projectFailure, mapOutput, isVerified? }` — `toolProvider` fills `projectFailure` with the standard HTTP projection; override it only for a provider whose failure shape is nonstandard (see `enrichments/provider-failures/`). Providers are **plain data** (no `@/tools` import) so the catalog stays client-safe.
 - **`enrichments/providers.ts`** — `toolProvider(...)` (typed passthrough) plus shared input helpers: `str(v)`, `normalizeDomain(v)`, `firstNonEmpty(arr)`, `splitName(fullName)`, and `projectEnrichmentProviderFailure`.
 - **`enrichments/run.ts`** — the server-only cascade runner. Calls `executeTool(provider.toolId, { ...params, _context: { workspaceId, userId } })`, accumulates hosted-key cost, returns the first non-empty mapped result. **You do not edit this** — it works for any registry entry.
 - **`enrichments/registry.ts`** — `ENRICHMENT_REGISTRY` / `ALL_ENRICHMENTS` / `getEnrichment`. Register new entries here.
@@ -37,7 +37,9 @@ For each output the enrichment produces, decide which existing tool provides it.
 - Its `params` accept what you can derive from table columns (read the tool's `params`).
 - Its `outputs` / `transformResponse` actually expose the field you need (read the real output shape — don't assume).
 
-Order providers **cheapest / most-likely-to-hit first**; the cascade stops at the first non-empty result. Apollo and LinkedIn APIs are not hosted-safe (ToS) — never call them as providers.
+Order providers by **accuracy first, then price per hit**. Order decides which answer wins when two providers would disagree, so a cheaper but less accurate provider ahead of a better one trades quality for cost; reorder for price only with benchmark evidence. Among similarly accurate providers, put the ones that bill on a miss or charge a high price per hit after the cheaper, free-on-miss ones (check each vendor's billing docs and the tool's `hosting.pricing`); never move a more accurate provider behind a less accurate one to save cost. The cascade stops at the first non-empty, verified result. Apollo and LinkedIn APIs are not hosted-safe (ToS) — never call them as providers.
+
+Prefer a vendor's own verified-only option (e.g. Prospeo `only_verified_email`) when it turns an unverified result into a free miss. Otherwise, when a provider can return an unconfirmed hit (a catch-all guess, a `risky` status), declare `isVerified(output)`: an unconfirmed hit is held while later providers look for a confirmed one, and fills the cell only if none does. `isVerified: false` marks a provider that never confirms its hits (a database record); it runs only while nothing is held.
 
 ## Step 2: Verify hosted-key support — chain to `/add-hosted-key` if missing
 
@@ -136,7 +138,8 @@ export const ENRICHMENT_REGISTRY: EnrichmentRegistry = {
 
 - [ ] Each output mapped to a real tool field (verified against the tool's `params`/`outputs`)
 - [ ] **Every provider tool has a `hosting` block — ran `/add-hosted-key` for any that didn't**
-- [ ] Providers ordered cheapest / most-likely-first; Apollo/LinkedIn not used
+- [ ] Providers ordered accuracy-first, then price per hit (billing on a miss checked against vendor docs); Apollo/LinkedIn not used
+- [ ] Providers that can return an unconfirmed hit declare `isVerified`
 - [ ] Enrichment file is client-safe (no `@/tools` import); uses `toolProvider` + shared helpers
 - [ ] `buildParams` returns `null` on insufficient inputs; `mapOutput` returns `null` on a miss
 - [ ] Registered in `enrichments/registry.ts`
 
@@ -33,7 +33,7 @@ contracts, and demonstrated regressions.
 | Suffix | Needs | Run with | In CI |
 |--------|-------|----------|-------|
 | `*.test.ts(x)` | nothing; global mocks from `vitest.setup.ts` | `bun run --cwd <workspace> test` | `test` jobs |
-| `*.integration.ts` | real PostgreSQL (`TEST_DATABASE_URL`), optionally Redis (`TEST_REDIS_URL`) | `bun run --cwd <workspace> test --mode integration` | `integration` jobs, by glob (sharded per provisioning path) |
+| `*.integration.ts` | real PostgreSQL (`TEST_DATABASE_URL`), Redis (`TEST_REDIS_URL`; optional locally, required in CI) | `bun run --cwd <workspace> test --mode integration` | `integration` jobs, by glob (sharded per provisioning path) |
 | `*.live.test.ts` | provider APIs, hosted sandboxes, local runtimes, or sibling checkouts | `bun run --cwd apps/sim test --mode live <file>` | never |
 | `apps/desktop/e2e/*.spec.ts` | the packaged Electron app | Playwright | desktop E2E workflow |
 | `apps/sim/scripts/test-*-e2e.ts` | a running app over HTTP | its `package.json` script when one exists (`bun run test:scim:e2e`; `test:workflow-version-compare:e2e` adds `--no-env-file`), else `bun scripts/test-<suite>-e2e.ts` from apps/sim | `e2e` jobs (`.github/scripts/http-e2e.sh`) |
@@ -48,8 +48,8 @@ contracts, and demonstrated regressions.
   those checks. Isolate with a unique schema or generated IDs, and clean up in `afterAll`.
 - Integration files run one at a time against one shared database. A new `*.integration.ts` is
   picked up by CI with no workflow change, and the run writes the JSON report configured in
-  `vitest.shared.ts`, which CI uploads. Never add a passing suite to the quarantine list in
-  `apps/sim/vitest.config.ts`.
+  `vitest.shared.ts`, which CI uploads. Never exclude a failing suite from the run in
+  `apps/sim/vitest.config.ts`: fix it or delete it.
 - `bun run test:integration` starts disposable Postgres and Redis containers, provisions the schema,
   and runs both workspaces; pass filenames to narrow the `apps/sim` run.
 
@@ -169,6 +169,7 @@ The suite's wall time is bound by the single Vite server thread that serves ever
 3. Mock heavy graphs a test does not need and the setup does not already mock: `@/blocks`,
    `@/triggers/registry`, `@/tools/generated/*`.
 4. No real timers: `vi.useFakeTimers()`, `flushMicrotasks()`, or `flushMacrotask()`.
+   `check:test-patterns` (`real-wait`) fails on a new wall-clock wait in a unit test.
 5. Absolute imports only.
 
 ### Running
 
@@ -31,7 +31,7 @@ contracts, and demonstrated regressions.
 | Suffix | Needs | Run with | In CI |
 |--------|-------|----------|-------|
 | `*.test.ts(x)` | nothing; global mocks from `vitest.setup.ts` | `bun run --cwd <workspace> test` | `test` jobs |
-| `*.integration.ts` | real PostgreSQL (`TEST_DATABASE_URL`), optionally Redis (`TEST_REDIS_URL`) | `bun run --cwd <workspace> test --mode integration` | `integration` jobs, by glob (sharded per provisioning path) |
+| `*.integration.ts` | real PostgreSQL (`TEST_DATABASE_URL`), Redis (`TEST_REDIS_URL`; optional locally, required in CI) | `bun run --cwd <workspace> test --mode integration` | `integration` jobs, by glob (sharded per provisioning path) |
 | `*.live.test.ts` | provider APIs, hosted sandboxes, local runtimes, or sibling checkouts | `bun run --cwd apps/sim test --mode live <file>` | never |
 | `apps/desktop/e2e/*.spec.ts` | the packaged Electron app | Playwright | desktop E2E workflow |
 | `apps/sim/scripts/test-*-e2e.ts` | a running app over HTTP | its `package.json` script when one exists (`bun run test:scim:e2e`; `test:workflow-version-compare:e2e` adds `--no-env-file`), else `bun scripts/test-<suite>-e2e.ts` from apps/sim | `e2e` jobs (`.github/scripts/http-e2e.sh`) |
@@ -46,8 +46,8 @@ contracts, and demonstrated regressions.
   those checks. Isolate with a unique schema or generated IDs, and clean up in `afterAll`.
 - Integration files run one at a time against one shared database. A new `*.integration.ts` is
   picked up by CI with no workflow change, and the run writes the JSON report configured in
-  `vitest.shared.ts`, which CI uploads. Never add a passing suite to the quarantine list in
-  `apps/sim/vitest.config.ts`.
+  `vitest.shared.ts`, which CI uploads. Never exclude a failing suite from the run in
+  `apps/sim/vitest.config.ts`: fix it or delete it.
 - `bun run test:integration` starts disposable Postgres and Redis containers, provisions the schema,
   and runs both workspaces; pass filenames to narrow the `apps/sim` run.
 
@@ -167,6 +167,7 @@ The suite's wall time is bound by the single Vite server thread that serves ever
 3. Mock heavy graphs a test does not need and the setup does not already mock: `@/blocks`,
    `@/triggers/registry`, `@/tools/generated/*`.
 4. No real timers: `vi.useFakeTimers()`, `flushMicrotasks()`, or `flushMacrotask()`.
+   `check:test-patterns` (`real-wait`) fails on a new wall-clock wait in a unit test.
 5. Absolute imports only.
 
 ### Running
 
@@ -0,0 +1,65 @@
+#!/usr/bin/env bash
+# Prints which checks a pull request's changes can affect, as `<flag>=true|false` lines for
+# $GITHUB_OUTPUT. A flag is `false` only when every changed file is on an allowlist whose jobs
+# fully cover it. A file on no allowlist sets every flag but `desktop_live` (which keeps its own
+# skip list) to `true`; a change under .github/, an empty diff, or any failure to work out the
+# diff sets all of them.
+#
+#   full          integration, e2e, e2e-build, e2e-mobile: any file not allowlisted below
+#   test          the Vitest shards and test-types: content, docs app, and desktop changes,
+#                 plus `full`
+#   build         the app's `next build`: content changes, plus `full`
+#   desktop_live  the live Electron suite: any file outside its own skip list
+#
+# lint always runs and is not gated.
+#
+# Usage: ci-changes.sh <base-sha>
+set -u
+
+# Posts, author profiles, and the public assets only posts reference. The test shards read them
+# (lib/content/og-image.test.ts) and `next build` prerenders every post with its MDX components,
+# which check:library-content compiles but never renders. Components under content/ are code.
+content='^apps/sim/content/([^/]+/)*[^/]+\.mdx?$|^apps/sim/content/authors/[^/]+\.json$|^apps/sim/public/(authors|blog|changelog|library)/'
+# Its own workspace tests, its type-check and docs-manifest audit in lint, and apps/sim tests
+# that read its pages (tools/netsuite/netsuite.test.ts).
+docs='^apps/docs/'
+# Nothing outside the Electron app imports it; the desktop bridge and protocol packages are
+# consumed by apps/sim and stay unlisted.
+desktop='^apps/desktop/'
+# Agent guidance, read only by the lint job's audits.
+guidance='^(\.agents|\.claude|\.cursor)/|^([^/]+/)*(AGENTS|CLAUDE)\.md$|^README\.md$'
+# The live desktop suite drives apps/desktop against the app, so it skips more than the rest.
+desktop_live_skip="$guidance"'|^(apps/docs/|apps/pii/|apps/sim/content/|apps/sim/public/(authors|blog|changelog|library)/|packages/(python-sdk|ts-sdk)/)|\.mdx?$|(^|/)LICENSE$'
+
+emit() {
+  printf 'full=%s\ntest=%s\nbuild=%s\ndesktop_live=%s\n' "$1" "$2" "$3" "$4"
+}
+run_all() {
+  emit true true true true
+  echo "Running every check: $1" >&2
+  exit 0
+}
+matches() { printf '%s\n' "$names" | grep -qE "$1"; }
+outside() { printf '%s\n' "$names" | grep -qvE "$1"; }
+
+base=${1:-}
+[ -n "$base" ] || run_all 'no base commit'
+git fetch --quiet --depth=1 origin "$base" || run_all "could not fetch $base"
+names=$(git diff --no-renames --name-only "$base" HEAD) || run_all "could not diff against $base"
+[ -n "$names" ] || run_all 'no changed files listed'
+if matches '^\.github/'; then run_all 'a CI file changed'; fi
+
+desktop_live=false
+if outside "$desktop_live_skip"; then desktop_live=true; fi
+if outside "$content|$docs|$desktop|$guidance"; then
+  emit true true true "$desktop_live"
+  unlisted=$(printf '%s\n' "$names" | grep -vE "$content|$docs|$desktop|$guidance" | head -n 1)
+  echo "Running every check (desktop_live=$desktop_live): $unlisted is not on an allowlist" >&2
+  exit 0
+fi
+
+test=false build=false
+if matches "$content|$docs|$desktop"; then test=true; fi
+if matches "$content"; then build=true; fi
+emit false "$test" "$build" "$desktop_live"
+echo "Scoped checks: test=$test build=$build desktop_live=$desktop_live" >&2
@@ -0,0 +1,40 @@
+#!/usr/bin/env bash
+# The `ci` job's verdict: the single required check, so it must never pass a run whose checks did
+# not all pass. Reads the job's `needs` context (`toJSON(needs)`) from $NEEDS and exits non-zero,
+# naming each offending job, unless every job succeeded or was skipped as `changes` intended.
+#
+# A skip passes only for a gated job whose `changes` flag is exactly `false`:
+#
+#   integration, e2e, e2e-build, e2e-mobile  full
+#   test, test-types                         test
+#   build                                    build
+#   desktop-live                             desktop_live
+#
+# Every other skip fails: an ungated job (lint, `changes` itself), a gated job whose flag is
+# `true`, missing, or anything else (a failed or skipped `changes` leaves no outputs). A failure
+# or a cancellation always fails.
+#
+# Usage: NEEDS='<toJSON(needs)>' ci-gate.sh
+set -euo pipefail
+
+needs=${NEEDS:-}
+if ! jq -e 'type == "object"' <<< "$needs" > /dev/null 2>&1; then
+  echo "::error::The needs context is not a JSON object: '$needs'"
+  exit 1
+fi
+
+failed="$(jq -r '
+  {"integration": "full", "e2e": "full", "e2e-build": "full", "e2e-mobile": "full", "test": "test",
+   "test-types": "test", "build": "build", "desktop-live": "desktop_live"} as $gate
+  | (.changes.outputs // {}) as $flags
+  | to_entries[]
+  | select(.value.result != "success")
+  | select(.value.result != "skipped" or $gate[.key] == null or $flags[$gate[.key]] != "false")
+  | "\(.key): \(.value.result)"
+' <<< "$needs")"
+if [ -n "$failed" ]; then
+  echo "::error::Checks did not pass:"
+  echo "$failed"
+  exit 1
+fi
+echo "All checks passed."