Added value: +{
+  "anyOf": [
+    {
+      "type": "string"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "An agency client's email, from list_agency_clients, to run this call in that client's account: their agents, settings, connected accounts, and credits. Omit it to act on your own account."
+}

Previous value: -"The company's website domain, e.g. 'acme.com'. Pass it whenever you have it — a domain names one employer, where a company name matches every company written that way, and same-name companies are the main source of wrong-company results. A company row's `identifier` holds one; pass it as stored, whether that is a bare host or a full URL."New value: +"The company's website domain, e.g. 'acme.com'. Pass it whenever you have it — a domain names one employer, where a company name matches every company written that way, and same-name companies are the main source of wrong-company results. A company row's `identifier` holds its domain; pass it as stored."

Removed value: -{
-  "default": true,
-  "description": "Default True — in an agent, checked people are saved and linked\nas person rows automatically (see `list_name`). Pass False only for a\nread-only lookup the user doesn't want on a list.",
-  "type": "boolean"
-}

Previous value: -"Optional — a specific agent to save the matched people into.\nOmit it to use the running agent, which is the usual case. Only a\nthrowaway lookup outside any agent returns results without saving."New value: +"Optional — a specific agent to save the checked people into.\nOmit it to use the running agent, which is the usual case. Only a\nthrowaway lookup outside any agent returns results without saving."

Previous value: -"Short kebab slug naming the Output-tab list bucket (e.g.\n'acme-execs'). When this runs in an agent, matched people are saved\nand linked as `agent_search_results` person rows automatically —\ndeduped by profile, re-runs update in place; `list_name` names\ntheir list, and absent it they land in the 'default' list. Query\nthem by reference with `query_search_results` instead of re-typing\nURLs from `results`. The `results` return is unchanged either way."New value: +"Short kebab slug naming the Output-tab list bucket (e.g.\n'acme-execs'). When this runs in an agent, the people the check\njudged are saved and linked as `agent_search_results` person rows\nautomatically, each with its verdict (the rejected ones hidden from\nthe list's default view) — deduped by profile, re-runs update in\nplace; `list_name` names their list, and absent it they land in\nthe 'default' list. Query\nthem by reference with `query_search_results` instead of re-typing\nURLs from `results`. The `results` return is unchanged either way."

Previous value: -"Candidates Exa retrieves and ranks per search before\nthe match filter runs. Default 10; max 100. Matches cluster at\nthe top of Exa's ranking, so 10 usually holds the right people\nfor a well-formed query; the returned `results` are shorter\nafter filtering. If `results` is thin and `rejected` is long,\nreword the query (the reject reasons say whether and how)\nrather than raising this — the tail ranks are lower-relevance\nanyway."New value: +"Candidates Exa retrieves and ranks per search before\nthe check runs. Default 10; max 100. Matches cluster at\nthe top of Exa's ranking, so 10 usually holds the right people\nfor a well-formed query; the returned `results` are shorter\nafter the check. If `results` is thin and `rejected` is long,\nreword the query (the reject reasons say whether and how)\nrather than raising this — the tail ranks are lower-relevance\nanyway."

Previous value: -"Default True — in an agent, matched people are saved and linked\nas person rows automatically (see `list_name`). Pass False only for a\nread-only lookup the user doesn't want on a list."New value: +"Default True — in an agent, checked people are saved and linked\nas person rows automatically (see `list_name`). Pass False only for a\nread-only lookup the user doesn't want on a list."

Previous value: -"List of `ExaPeopleSearch` objects. Each has its own\n`query` + optional `system_prompt`. For a single-company\nlookup, pass a list of one."New value: +"List of `ExaPeopleSearch` objects. Each has its own\n`query` + `criteria`. For a single-company lookup, pass a list\nof one."

Previous value: -"One people search: who to find, and at which company.\n\nMultiple of these can be batched into a single `exa_find_people`\ncall — the tool fans them out in parallel and returns results in\ninput order. Each search is independent (its own query +\nsystem_prompt); the shared `num_results` is passed as a kwarg on the\nparent tool call."New value: +"One people search: who to find, and at which company.\n\nMultiple of these can be batched into a single `exa_find_people`\ncall — the tool fans them out in parallel and returns results in\ninput order. Each search is independent (its own query +\ncriteria); the shared `num_results` is passed as a kwarg on the\nparent tool call."

Previous value: -"The company this search is aimed at, by name, on its own. Anyone placed at a differently-named employer is then dropped before you see them — a founder of 'Acmely' coming back for an 'Acme' search. The company name inside `query` / `system_prompt` steers ranking; this is what filters. Required, with no default, so a search running unfiltered is a decision rather than an omission: pass null for a cross-company search ('pricing leaders across chemical distributors'), and for an alumni search, where the person's current employer is a different company by definition."New value: +"The company this search is aimed at, by name, on its own; null for a cross-company search ('pricing leaders across chemical distributors') or an alumni search. When `company_domain` / `company_linkedin_url` pins the search, the check is also told this name and that identifier as the employer searched for, so it can tell that employer's staff from a same-named company's. Requiring that they work there now is still `criteria`'s job."

Added value: +{
+  "description": "What a match is for this search, in plain words: the requirements a person must meet, led by the employer and that they work there now (or the one person you're after), then any role, seniority, region or exclusion that decides it. The right place for exclusions. Each person returned is checked against all of it as requirements, so leave out a preference that only orders results (VPs ahead of Directors, the founder ahead of a first sales hire) and name everyone you'd accept. Example: 'Currently works at Microsoft, not a former employee. Product leader. Not a growth marketer.'",
+  "minLength": 1,
+  "type": "string"
+}

Previous value: -"Places the person must be based in, by country, state or city name, e.g. ['United States'] or ['Texas', 'Colorado']. Entries are OR-ed, and the filter applies whichever provider serves the search. Spell states out: a two-letter code is read as a country, so 'CT' does not mean Connecticut. At a multinational this is what keeps overseas staff out of the results — naming a place in `query` or `system_prompt` does not filter a search pinned by `company_domain` / `company_linkedin_url`."New value: +"Places the person must be based in, by country, state or city name, e.g. ['United States'] or ['Texas', 'Colorado']. Entries are OR-ed, and the filter applies whichever provider serves the search. Spell states out: a two-letter code is read as a country, so 'CT' does not mean Connecticut. At a multinational this is what keeps overseas staff out of what a search pinned by `company_domain` / `company_linkedin_url` fetches: `criteria` only rejects from what was fetched, and `query` doesn't reach that search at all."

Previous value: -"POSITIVE-FORM description of who you want, including the company name. Examples: 'VP of Product at Microsoft', 'senior marketing leaders at Exaforce'. Do NOT include exclusions here — vector retrieval has no concept of 'not', so naming an excluded term biases the ranker TOWARDS it."New value: +"POSITIVE-FORM description of who you want, including the company name. Examples: 'VP of Product at Microsoft', 'senior marketing leaders at Exaforce'. On a search not pinned by `company_domain` / `company_linkedin_url`, include the region the ask names ('VP of Product at Microsoft in the US'): that search has no location filter, so `locations` and `criteria` only reject from the people `query` retrieves. Do NOT include exclusions here — vector retrieval has no concept of 'not', so naming an excluded term biases the ranker TOWARDS it."

Removed value: -{
-  "anyOf": [
-    {
-      "type": "string"
-    },
-    {
-      "type": "null"
-    }
-  ],
-  "default": null,
-  "description": "Optional ranker strategy hint — the right place for preferences and exclusions. Example: 'MUST currently work at Microsoft. Prioritize VP+ over Director+. Exclude growth marketers.'"
-}

Previous value: -"Role keywords, e.g. ['Operations', 'Training']. Each is matched as a case-insensitive substring of the person's job title and the list is OR-ed, so short function or seniority words are what match: 'Marketing', 'Revenue', 'Director', 'Chief'. A full title phrase matches only where it appears verbatim, which misses every company that words that title differently or abbreviates it — so always include the broad function word on its own ('Marketing') before adding specific titles, and pass both spellings of a rank ('VP' and 'Vice President' match different people). Entries are OR-ed, so a title that matches nobody costs nothing. Narrowing also decides WHICH people come back at a big employer: results are unordered, so without it you get an arbitrary slice of everyone there rather than the roles you asked for."New value: +"Role keywords, e.g. ['Operations', 'Training']. Each is matched as a case-insensitive substring of the person's job title and the list is OR-ed, so short function or seniority words are what match: 'Marketing', 'Revenue', 'Director', 'Chief'. A full title phrase matches only where it appears verbatim, which misses every company that words that title differently or abbreviates it — so always include the broad function word on its own ('Marketing') before adding specific titles, and pass both spellings of a rank ('VP' and 'Vice President' match different people). Entries are OR-ed, so a title that matches nobody costs nothing. Narrowing also decides WHICH people come back at a big employer: the fetch is unordered, so without it you get an arbitrary slice of everyone there rather than the roles you asked for."

Previous value: -[
-  "query",
-  "company"
-]New value: +[
+  "query",
+  "criteria",
+  "company"
+]

Previous value: -"Candidates Exa retrieves and ranks per search before\nthe match filter runs. Default 10; max 100. Matches cluster at\nthe top of Exa's ranking, so 10 usually holds the right people\nfor a well-formed query; the returned `results` are shorter\nafter filtering. If `results` is thin and `rejected` is long,\nreword the query (the reject reasons say how) rather than\nraising this — the tail ranks are lower-relevance anyway."New value: +"Candidates Exa retrieves and ranks per search before\nthe match filter runs. Default 10; max 100. Matches cluster at\nthe top of Exa's ranking, so 10 usually holds the right people\nfor a well-formed query; the returned `results` are shorter\nafter filtering. If `results` is thin and `rejected` is long,\nreword the query (the reject reasons say whether and how)\nrather than raising this — the tail ranks are lower-relevance\nanyway."

Previous value: -"Default True — in an agent, matched people are saved and linked\nas person rows automatically (see `list_name`). Pass False to return\nresults without saving, for a flow that folds these people into\nanother row instead — e.g. a company-find that nests them under each\ncompany row via `record_search_results`, where a standalone person\nlist would duplicate people already shown under their company."New value: +"Default True — in an agent, matched people are saved and linked\nas person rows automatically (see `list_name`). Pass False only for a\nread-only lookup the user doesn't want on a list."

Added value: +{
+  "anyOf": [
+    {
+      "items": {
+        "type": "string"
+      },
+      "type": "array"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Places the person must be based in, by country, state or city name, e.g. ['United States'] or ['Texas', 'Colorado']. Entries are OR-ed, and the filter applies whichever provider serves the search. Spell states out: a two-letter code is read as a country, so 'CT' does not mean Connecticut. At a multinational this is what keeps overseas staff out of the results — naming a place in `query` or `system_prompt` does not filter a search pinned by `company_domain` / `company_linkedin_url`."
+}

Added value: +{
+  "anyOf": [
+    {
+      "type": "integer"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Optional — a specific agent to save the matched people into.\nOmit it to use the running agent, which is the usual case. Only a\nthrowaway lookup outside any agent returns results without saving."
+}

Previous value: -"Short kebab slug naming the Output-tab list bucket (e.g.\n'acme-execs'). When this runs in a task, matched people are saved\nand linked as `agent_search_results` person rows automatically —\ndeduped by profile, re-runs update in place; `list_name` names\ntheir list, and absent it they land in the 'default' list. Query\nthem by reference with `query_search_results` instead of re-typing\nURLs from `results`. The `results` return is unchanged either way."New value: +"Short kebab slug naming the Output-tab list bucket (e.g.\n'acme-execs'). When this runs in an agent, matched people are saved\nand linked as `agent_search_results` person rows automatically —\ndeduped by profile, re-runs update in place; `list_name` names\ntheir list, and absent it they land in the 'default' list. Query\nthem by reference with `query_search_results` instead of re-typing\nURLs from `results`. The `results` return is unchanged either way."

Previous value: -"Default True — in a task, matched people are saved and linked\nas person rows automatically (see `list_name`). Pass False to return\nresults without saving, for a flow that folds these people into\nanother row instead — e.g. a company-find that nests them under each\ncompany row via `record_search_results`, where a standalone person\nlist would duplicate people already shown under their company."New value: +"Default True — in an agent, matched people are saved and linked\nas person rows automatically (see `list_name`). Pass False to return\nresults without saving, for a flow that folds these people into\nanother row instead — e.g. a company-find that nests them under each\ncompany row via `record_search_results`, where a standalone person\nlist would duplicate people already shown under their company."

Removed value: -{
-  "anyOf": [
-    {
-      "type": "integer"
-    },
-    {
-      "type": "null"
-    }
-  ],
-  "default": null,
-  "description": "Optional — a specific task to save the matched people into.\nOmit it to use the running task, which is the usual case. Only a\nthrowaway lookup outside any task returns results without saving."
-}

Previous value: -"Short kebab slug naming the list bucket (e.g.\n'acme-execs'). Pass together with `task_id`. When set, the\nmatched people are also saved as `agent_search_results` person\nrows under this slug (deduped by profile, re-runs update in\nplace) — query them by reference with `query_search_results`\ninstead of re-typing URLs from `results`. The `results` return\nis unchanged either way."New value: +"Short kebab slug naming the Output-tab list bucket (e.g.\n'acme-execs'). When this runs in a task, matched people are saved\nand linked as `agent_search_results` person rows automatically —\ndeduped by profile, re-runs update in place; `list_name` names\ntheir list, and absent it they land in the 'default' list. Query\nthem by reference with `query_search_results` instead of re-typing\nURLs from `results`. The `results` return is unchanged either way."

Added value: +{
+  "default": true,
+  "description": "Default True — in a task, matched people are saved and linked\nas person rows automatically (see `list_name`). Pass False to return\nresults without saving, for a flow that folds these people into\nanother row instead — e.g. a company-find that nests them under each\ncompany row via `record_search_results`, where a standalone person\nlist would duplicate people already shown under their company.",
+  "type": "boolean"
+}

Previous value: -"Pass together with `list_name` to save every matched\nperson into this task's Output-tab list. Omit both for a\nthrowaway lookup."New value: +"Optional — a specific task to save the matched people into.\nOmit it to use the running task, which is the usual case. Only a\nthrowaway lookup outside any task returns results without saving."