diff --git a/scripts/check-skills-token-ratchet.mjs b/scripts/check-skills-token-ratchet.mjs
index db0e70145b..c36f89f042 100644
--- a/scripts/check-skills-token-ratchet.mjs
+++ b/scripts/check-skills-token-ratchet.mjs
@@ -425,7 +425,6 @@ export const CEILINGS = new Map([
   ['skills/objectstack-platform/rules/service-registry.md', 2331],
 
   // objectstack-query
-  ['skills/objectstack-query/evals/README.md', 567],
   ['skills/objectstack-query/rules/aggregation.md', 2357],
   ['skills/objectstack-query/rules/filters.md', 2149],
   ['skills/objectstack-query/rules/pagination.md', 1382],
diff --git a/skills/objectstack-query/evals/README.md b/skills/objectstack-query/evals/README.md
deleted file mode 100644
index 58f0f20c4a..0000000000
--- a/skills/objectstack-query/evals/README.md
+++ /dev/null
@@ -1,38 +0,0 @@
-# objectstack-query — Evals
-
-Test cases for the objectstack-query skill. Each eval gives the model a
-natural-language query need and checks the emitted ObjectQL against the
-subset the engine actually executes.
-
-## Planned Evals
-
-1. **Operator choice** — "Filter accounts where status is active or pending,
-   excluding deleted ones." Expect `$in` (or `$or`) plus `$ne`/`$not`;
-   penalize string operators on non-string fields and `deleted_at: null`
-   instead of `$null`.
-2. **Nested relation filter** — "Find orders where the customer's country is
-   US." Expect a nested relation filter (`customer: { country: 'US' }`),
-   not a `joins` array (removed in protocol 17).
-3. **Pagination pattern** — "Implement infinite scroll for a feed." Expect
-   manual keyset pagination (`where` on the sort key + `orderBy` + `limit`);
-   fail if the answer uses the removed `cursor` property.
-4. **Aggregation correctness** — "Count deals by region and show total
-   revenue." Expect `groupBy` + `count`/`sum` with aliases; fail on a missing
-   `alias`, or on `array_agg`/`string_agg` (removed in protocol 17).
-5. **Filtered aggregation** — "One call: total orders and count of orders
-   over $1000." Expect one call with a per-aggregation `filter` on the
-   conditional measure; fail on `where`, which scopes every measure.
-6. **Post-aggregation filtering** — "Customers with more than 5 orders."
-   Expect `having` on the aggregation alias; fail on `where`, which filters
-   input rows before the alias exists.
-7. **Date-bucketed time series** — "Monthly revenue for the last year."
-   Expect structured `groupBy` with `dateGranularity: 'month'`, not
-   client-side bucketing and not `windowFunctions` (removed in protocol 17).
-8. **Expand vs direct query** — "Show a task list with assignee names; page
-   through one project's tasks." Expect `expand` for the lookup display and
-   a direct query on the related object for pagination (nested
-   limit/offset are not applied inside `expand`).
-9. **Search subset** — "Keyword search over articles." Expect
-   `search: { query, fields }` only; fail if the answer relies on `fuzzy`,
-   `boost`, `operator`, `minScore`, `language`, or `highlight` (all
-   schema-reserved).