From d40e110e53919bca024cf491023ed87711ec51e9 Mon Sep 17 00:00:00 2001 From: Vlad Shulman Date: Fri, 12 Jun 2026 14:01:17 -0700 Subject: [PATCH] Note recall optimization in findall entity-search help text Aligns the command help with the docs framing (https://docs.parallel.ai/findall-api/entity-search): results are ranked but not individually verified; findall run remains the path for comprehensive match evaluation. Co-Authored-By: Claude Fable 5 --- parallel_web_tools/cli/commands.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/parallel_web_tools/cli/commands.py b/parallel_web_tools/cli/commands.py index 2467ac2..45e3208 100644 --- a/parallel_web_tools/cli/commands.py +++ b/parallel_web_tools/cli/commands.py @@ -2787,7 +2787,8 @@ def findall_entity_search( ): """Return ranked entities matching a natural language objective. - A best-effort, low-latency search. For comprehensive match evaluation and + A best-effort, low-latency search optimized for recall: results are ranked + but not individually verified. For comprehensive match evaluation and enrichment, use `parallel-cli findall run` instead. OBJECTIVE is a natural language description of what to find.