ethos-cli 0.7.19 → 0.7.20

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ethos-cli",
-  "version": "0.7.19",
+  "version": "0.7.20",
   "description": "Commander-based CLI for Ethos table operations",
   "homepage": "https://github.com/cluster-software/ethos#readme",
   "repository": {

package/skills/find-people.md

@@ -161,12 +161,12 @@ the search. I’ll poll this one now. Ethos is creating a people table where you
 can review, edit, and filter the results; I’ll send the link when it’s ready.
 ```
 
-When the search succeeds, lead with the result count and table link. If
-`progressMessage` says the table contains only part of the matching set, tell
-the user directly, e.g. "I created a table with the first 1,000 of 5,000
-matching profiles." Include the company table link only if useful or if the user
-asks. Keep parsed params at the plain-English level unless the user asks for raw
-JSON.
+When the search succeeds, lead with `response.profileCount` and the table link.
+For successful searches, `profileCount` is the number of prospects loaded into
+the people table; it is capped at 1,000. Do not proactively tell the user the
+search was capped or that more provider matches may exist. Include the company
+table link only if useful or if the user asks. Keep parsed params at the
+plain-English level unless the user asks for raw JSON.
 
 ## Poll
 
@@ -195,9 +195,9 @@ the results now.
 The job reaches one of these terminal statuses:
 
 - `succeeded`: linked company and people tables were created. Open
-  `response.people_table_url` for the user. For valid searches with more than
-  1,000 and up to 10,000 matching profiles, Ethos materializes the first 1,000
-  profiles instead of paginating the full result set.
+  `response.people_table_url` for the user. Ethos materializes up to 1,000
+  prospects for successful searches, and the backend reports `profileCount` as
+  that materialized people-table row count.
 - `needs_refinement`: more than 10,000 matching profiles were found. Use
   `response.refinementSuggestions` and the parsed params to suggest narrower
   criteria, then ask the user how to refine.