From 1c9e8d21c09a41df69cb1ca45d06d31881f4e66a Mon Sep 17 00:00:00 2001
From: lorenzejay <lorenzejaytech@gmail.com>
Date: Tue, 7 Apr 2026 17:49:06 -0700
Subject: [PATCH 1/3] feat: introduce MemoryPromptConfig for customizable
 memory prompts

* Added MemoryPromptConfig class to allow users to override default memory prompts for various operations (save, query, extract, consolidation).
* Updated relevant functions and classes to utilize the new configuration, enabling more flexible and context-specific memory handling.
* Enhanced tests to validate the functionality of the new prompt configuration and its integration within the memory processing flows.
---
 lib/crewai/src/crewai/__init__.py             |   2 +
 lib/crewai/src/crewai/memory/analyze.py       |  63 ++++++---
 lib/crewai/src/crewai/memory/encoding_flow.py |   4 +
 lib/crewai/src/crewai/memory/recall_flow.py   |   1 +
 lib/crewai/src/crewai/memory/types.py         | 127 +++++++++++++++++-
 .../src/crewai/memory/unified_memory.py       |  25 +++-
 lib/crewai/src/crewai/telemetry/telemetry.py  |   7 +-
 lib/crewai/src/crewai/telemetry/utils.py      |  13 ++
 .../tests/memory/test_unified_memory.py       |  47 +++++++
 lib/crewai/tests/telemetry/test_telemetry.py  |  20 ++-
 10 files changed, 285 insertions(+), 24 deletions(-)

diff --git a/lib/crewai/src/crewai/__init__.py b/lib/crewai/src/crewai/__init__.py
index e82b925113..ab47aa6340 100644
--- a/lib/crewai/src/crewai/__init__.py
+++ b/lib/crewai/src/crewai/__init__.py
@@ -81,6 +81,7 @@ def _track_install_async() -> None:
 
 _LAZY_IMPORTS: dict[str, tuple[str, str]] = {
     "Memory": ("crewai.memory.unified_memory", "Memory"),
+    "MemoryPromptConfig": ("crewai.memory.types", "MemoryPromptConfig"),
 }
 
 
@@ -210,6 +211,7 @@ def __getattr__(name: str) -> Any:
     "Knowledge",
     "LLMGuardrail",
     "Memory",
+    "MemoryPromptConfig",
     "PlanningConfig",
     "Process",
     "RuntimeState",
diff --git a/lib/crewai/src/crewai/memory/analyze.py b/lib/crewai/src/crewai/memory/analyze.py
index e700f4281f..2709c57fcf 100644
--- a/lib/crewai/src/crewai/memory/analyze.py
+++ b/lib/crewai/src/crewai/memory/analyze.py
@@ -8,7 +8,7 @@
 
 from pydantic import BaseModel, ConfigDict, Field
 
-from crewai.memory.types import MemoryRecord, ScopeInfo
+from crewai.memory.types import MemoryPromptConfig, MemoryRecord, ScopeInfo
 from crewai.utilities.i18n import get_i18n
 
 
@@ -140,19 +140,23 @@ class ConsolidationPlan(BaseModel):
     )
 
 
-def _get_prompt(key: str) -> str:
-    """Retrieve a memory prompt from the i18n translations.
-
-    Args:
-        key: The prompt key under the "memory" section.
-
-    Returns:
-        The prompt string.
-    """
+def _memory_prompt_line(
+    memory_prompt: MemoryPromptConfig | None,
+    key: str,
+) -> str:
+    """Resolve one memory prompt: override string or bundled translation."""
+    if memory_prompt is not None:
+        raw = getattr(memory_prompt, key, None)
+        if isinstance(raw, str) and raw.strip():
+            return raw
     return get_i18n().memory(key)
 
 
-def extract_memories_from_content(content: str, llm: Any) -> list[str]:
+def extract_memories_from_content(
+    content: str,
+    llm: Any,
+    memory_prompt: MemoryPromptConfig | None = None,
+) -> list[str]:
     """Use the LLM to extract discrete memory statements from raw content.
 
     This is a pure helper: it does NOT store anything. Callers should call
@@ -164,15 +168,21 @@ def extract_memories_from_content(content: str, llm: Any) -> list[str]:
     Args:
         content: Raw text (e.g. task description + result dump).
         llm: The LLM instance to use.
+        memory_prompt: Optional per-step prompt strings (see ``MemoryPromptConfig``).
 
     Returns:
         List of short, self-contained memory statements (or [content] on failure).
     """
     if not (content or "").strip():
         return []
-    user = _get_prompt("extract_memories_user").format(content=content)
+    user = _memory_prompt_line(memory_prompt, "extract_memories_user").format(
+        content=content
+    )
     messages = [
-        {"role": "system", "content": _get_prompt("extract_memories_system")},
+        {
+            "role": "system",
+            "content": _memory_prompt_line(memory_prompt, "extract_memories_system"),
+        },
         {"role": "user", "content": user},
     ]
     try:
@@ -202,6 +212,7 @@ def analyze_query(
     available_scopes: list[str],
     scope_info: ScopeInfo | None,
     llm: Any,
+    memory_prompt: MemoryPromptConfig | None = None,
 ) -> QueryAnalysis:
     """Use the LLM to analyze a recall query.
 
@@ -212,6 +223,7 @@ def analyze_query(
         available_scopes: Scope paths that exist in the store.
         scope_info: Optional info about the current scope.
         llm: The LLM instance to use.
+        memory_prompt: Optional per-step prompt strings.
 
     Returns:
         QueryAnalysis with keywords, suggested_scopes, complexity, recall_queries, time_filter.
@@ -219,13 +231,16 @@ def analyze_query(
     scope_desc = ""
     if scope_info:
         scope_desc = f"Current scope has {scope_info.record_count} records, categories: {scope_info.categories}"
-    user = _get_prompt("query_user").format(
+    user = _memory_prompt_line(memory_prompt, "query_user").format(
         query=query,
         available_scopes=available_scopes or ["/"],
         scope_desc=scope_desc,
     )
     messages = [
-        {"role": "system", "content": _get_prompt("query_system")},
+        {
+            "role": "system",
+            "content": _memory_prompt_line(memory_prompt, "query_system"),
+        },
         {"role": "user", "content": user},
     ]
     try:
@@ -269,6 +284,7 @@ def analyze_for_save(
     existing_scopes: list[str],
     existing_categories: list[str],
     llm: Any,
+    memory_prompt: MemoryPromptConfig | None = None,
 ) -> MemoryAnalysis:
     """Infer scope, categories, importance, and metadata for a single memory.
 
@@ -280,17 +296,21 @@ def analyze_for_save(
         existing_scopes: Current scope paths in the memory store.
         existing_categories: Current categories in use.
         llm: The LLM instance to use.
+        memory_prompt: Optional per-step prompt strings.
 
     Returns:
         MemoryAnalysis with suggested_scope, categories, importance, extracted_metadata.
     """
-    user = _get_prompt("save_user").format(
+    user = _memory_prompt_line(memory_prompt, "save_user").format(
         content=content,
         existing_scopes=existing_scopes or ["/"],
         existing_categories=existing_categories or [],
     )
     messages = [
-        {"role": "system", "content": _get_prompt("save_system")},
+        {
+            "role": "system",
+            "content": _memory_prompt_line(memory_prompt, "save_system"),
+        },
         {"role": "user", "content": user},
     ]
     try:
@@ -322,6 +342,7 @@ def analyze_for_consolidation(
     new_content: str,
     existing_records: list[MemoryRecord],
     llm: Any,
+    memory_prompt: MemoryPromptConfig | None = None,
 ) -> ConsolidationPlan:
     """Decide insert/update/delete for a single memory against similar existing records.
 
@@ -332,6 +353,7 @@ def analyze_for_consolidation(
         new_content: The new content to store.
         existing_records: Existing records that are semantically similar.
         llm: The LLM instance to use.
+        memory_prompt: Optional per-step prompt strings.
 
     Returns:
         ConsolidationPlan with actions per record and whether to insert the new content.
@@ -345,12 +367,15 @@ def analyze_for_consolidation(
             f"- id={r.id} | scope={r.scope} | importance={r.importance:.2f} | created={created}\n"
             f"  content: {r.content[:200]}{'...' if len(r.content) > 200 else ''}"
         )
-    user = _get_prompt("consolidation_user").format(
+    user = _memory_prompt_line(memory_prompt, "consolidation_user").format(
         new_content=new_content,
         records_summary="\n\n".join(records_lines),
     )
     messages = [
-        {"role": "system", "content": _get_prompt("consolidation_system")},
+        {
+            "role": "system",
+            "content": _memory_prompt_line(memory_prompt, "consolidation_system"),
+        },
         {"role": "user", "content": user},
     ]
     try:
diff --git a/lib/crewai/src/crewai/memory/encoding_flow.py b/lib/crewai/src/crewai/memory/encoding_flow.py
index acd025d553..baf94c9c97 100644
--- a/lib/crewai/src/crewai/memory/encoding_flow.py
+++ b/lib/crewai/src/crewai/memory/encoding_flow.py
@@ -314,6 +314,7 @@ def parallel_analyze(self) -> None:
                         item.content,
                         list(item.similar_records),
                         self._llm,
+                        self._config.memory_prompt,
                     )
                 elif not fields_provided and not has_similar:
                     # Group C: field resolution only
@@ -324,6 +325,7 @@ def parallel_analyze(self) -> None:
                         existing_scopes,
                         existing_categories,
                         self._llm,
+                        self._config.memory_prompt,
                     )
                 else:
                     # Group D: both in parallel
@@ -334,6 +336,7 @@ def parallel_analyze(self) -> None:
                         existing_scopes,
                         existing_categories,
                         self._llm,
+                        self._config.memory_prompt,
                     )
                     consol_futures[i] = pool.submit(
                         contextvars.copy_context().run,
@@ -341,6 +344,7 @@ def parallel_analyze(self) -> None:
                         item.content,
                         list(item.similar_records),
                         self._llm,
+                        self._config.memory_prompt,
                     )
 
             # Collect field-resolution results
diff --git a/lib/crewai/src/crewai/memory/recall_flow.py b/lib/crewai/src/crewai/memory/recall_flow.py
index 3a058f27bd..3bc0b1a54f 100644
--- a/lib/crewai/src/crewai/memory/recall_flow.py
+++ b/lib/crewai/src/crewai/memory/recall_flow.py
@@ -227,6 +227,7 @@ def analyze_query_step(self) -> QueryAnalysis:
                 available,
                 scope_info,
                 self._llm,
+                self._config.memory_prompt,
             )
             self.state.query_analysis = analysis
 
diff --git a/lib/crewai/src/crewai/memory/types.py b/lib/crewai/src/crewai/memory/types.py
index e787b569d0..174ca2bf11 100644
--- a/lib/crewai/src/crewai/memory/types.py
+++ b/lib/crewai/src/crewai/memory/types.py
@@ -6,7 +6,7 @@
 from typing import Any
 from uuid import uuid4
 
-from pydantic import BaseModel, Field
+from pydantic import BaseModel, ConfigDict, Field
 
 
 # When searching the vector store, we ask for more results than the caller
@@ -132,6 +132,126 @@ class ScopeInfo(BaseModel):
     )
 
 
+class MemoryPromptConfig(BaseModel):
+    """Configuration for memory LLM prompts (like ``PlanningConfig`` for planning).
+
+    Field names match translation keys under ``memory`` in ``translations/en.json``.
+    When set, the string replaces the bundled prompt for that step; omitted keys
+    keep the default i18n text. Templates must include the same ``str.format``
+    placeholders as the defaults (e.g. ``save_user`` uses ``{content}``,
+    ``{existing_scopes}``, ``{existing_categories}``).
+    """
+
+    model_config = ConfigDict(extra="forbid")
+
+    save_system: str | None = None
+    save_user: str | None = None
+    query_system: str | None = None
+    query_user: str | None = None
+    extract_memories_system: str | None = None
+    extract_memories_user: str | None = None
+    consolidation_system: str | None = None
+    consolidation_user: str | None = None
+
+    @classmethod
+    def for_online_people_research(cls) -> MemoryPromptConfig:
+        """Prompt profile for open-web people research (e.g. Exa).
+
+        Biases memory toward **search provenance** (queries, domains, URLs, dead
+        ends, strong sources) and away from storing fragile biographical detail
+        or speculation as high-confidence fact. Pair with ``EXASearchTool`` and
+        explicit scopes/categories in ``remember()`` when you need determinism.
+        """
+        return cls(
+            save_system=(
+                "You analyze content for a hierarchical memory used during "
+                "online people research.\n"
+                "Prioritize durable *research-process* signals over raw biographical claims:\n"
+                "- suggested_scope: use paths like /subjects/<slug>, /searches/queries, "
+                "/sources/domains when the content is about where or how you searched.\n"
+                "- categories: include tags such as search_query, result_domain, "
+                "source_url, exa_search, dead_end, preferred_source, methodology "
+                "when applicable. Use subject_profile only for well-sourced factual "
+                "snippets tied to a named URL or document - not for gossip or guesses.\n"
+                "- importance: higher (0.7-1.0) for reusable search lessons "
+                "(e.g. 'site X returned 404', 'LinkedIn useful for role Y', "
+                "'query phrasing Z worked'). Lower (0.2-0.5) for one-off personal "
+                "minutiae or unverified claims.\n"
+                "- extracted_metadata.entities: prefer domains, publication names, "
+                "and search keywords; only list people when the text explicitly "
+                "anchors a fact to a source.\n"
+                "Given the content and existing scopes and categories, output:\n"
+                "1. suggested_scope\n2. categories\n3. importance\n4. extracted_metadata"
+            ),
+            save_user=(
+                "Content to store:\n{content}\n\n"
+                "Existing scopes: {existing_scopes}\n"
+                "Existing categories: {existing_categories}\n\n"
+                "Return the analysis as structured output."
+            ),
+            query_system=(
+                "You analyze a query against memory for people-research crews.\n"
+                "Favor recalling *how and where* the team already searched:\n"
+                "- keywords: include domains, site names, query stems, and subject "
+                "handles - not only the person's name.\n"
+                "- suggested_scopes: prefer /searches, /sources, /subjects as relevant.\n"
+                "- recall_queries: produce phrases that would match memories about "
+                "Exa queries, domains tried, and conclusions about source quality.\n"
+                "- complexity: 'complex' if the question asks to synthesize many sources; "
+                "else 'simple'.\n"
+                "- time_filter: ISO date if the query implies a time window; else null.\n"
+                "Given the query and available scopes, output:\n"
+                "1. keywords\n2. suggested_scopes\n3. complexity\n4. recall_queries\n"
+                "5. time_filter"
+            ),
+            query_user=(
+                "Query: {query}\n\n"
+                "Available scopes: {available_scopes}\n"
+                "{scope_desc}\n\n"
+                "Return the analysis as structured output."
+            ),
+            extract_memories_system=(
+                "Extract discrete memory statements from research logs (tool outputs, "
+                "Exa results, agent notes).\n"
+                "HIGH priority - always capture when present:\n"
+                "- Exact or paraphrased search queries used and which engine/tool "
+                "(e.g. Exa).\n"
+                "- Domains and canonical URLs of pages opened; HTTP errors or empty results.\n"
+                "- Judgments about source usefulness (e.g. 'official bio on company site', "
+                "'social profile - verify elsewhere').\n"
+                "MEDIUM priority:\n"
+                "- Verifiable facts explicitly tied to a source in the text "
+                "(quote which source).\n"
+                "LOW priority / often skip:\n"
+                "- Speculation, reputation gossip, or detailed private life unless "
+                "the task explicitly requires it and a primary source is named.\n"
+                "Each memory: one clear sentence, no duplicate ideas.\n"
+                'Output JSON with a single key "memories" whose value is a list of strings.'
+            ),
+            extract_memories_user=(
+                "Content:\n{content}\n\n"
+                "Extract memory statements as described. Return structured output."
+            ),
+            consolidation_system=(
+                "You consolidate new research content with similar existing memories.\n"
+                "Prefer keeping separate memories when they differ by URL, domain, or "
+                "search query - even if about the same person.\n"
+                "Merge or update when the new text clearly refines the same source "
+                "or same search finding.\n"
+                "Delete only when a memory is clearly wrong and superseded.\n"
+                "Be conservative: prefer 'keep' when unsure.\n"
+                "For each existing memory: keep | update | delete. "
+                "Set insert_new when the new content adds a distinct query, URL, or "
+                "source judgment."
+            ),
+            consolidation_user=(
+                "New content to consider storing:\n{new_content}\n\n"
+                "Existing similar memories:\n{records_summary}\n\n"
+                "Return the consolidation plan as structured output."
+            ),
+        )
+
+
 class MemoryConfig(BaseModel):
     """Internal configuration for memory scoring, consolidation, and recall behavior.
 
@@ -141,6 +261,11 @@ class MemoryConfig(BaseModel):
     compute_composite_score.
     """
 
+    memory_prompt: MemoryPromptConfig | None = Field(
+        default=None,
+        description="Per-step prompt strings overriding bundled memory prompts.",
+    )
+
     # -- Composite score weights --
     # The recall composite score is:
     #   semantic_weight * similarity + recency_weight * decay + importance_weight * importance
diff --git a/lib/crewai/src/crewai/memory/unified_memory.py b/lib/crewai/src/crewai/memory/unified_memory.py
index d879bace0c..bfc1eec941 100644
--- a/lib/crewai/src/crewai/memory/unified_memory.py
+++ b/lib/crewai/src/crewai/memory/unified_memory.py
@@ -9,7 +9,13 @@
 import time
 from typing import TYPE_CHECKING, Annotated, Any, Literal
 
-from pydantic import BaseModel, ConfigDict, Field, PlainValidator, PrivateAttr
+from pydantic import (
+    BaseModel,
+    ConfigDict,
+    Field,
+    PlainValidator,
+    PrivateAttr,
+)
 
 from crewai.events.event_bus import crewai_event_bus
 from crewai.events.types.memory_events import (
@@ -26,6 +32,7 @@
 from crewai.memory.types import (
     MemoryConfig,
     MemoryMatch,
+    MemoryPromptConfig,
     MemoryRecord,
     ScopeInfo,
     compute_composite_score,
@@ -59,6 +66,10 @@ class Memory(BaseModel):
     Works without agent/crew. Uses LLM to infer scope, categories, importance on save.
     Uses RecallFlow for adaptive-depth recall. Supports scope/slice views and
     pluggable storage (LanceDB default).
+
+    Override LLM prompts per step via ``memory_prompt`` (same idea as
+    ``PlanningConfig.system_prompt`` / ``plan_prompt``): set only the strings you
+    need; the rest stay on bundled translations.
     """
 
     model_config = ConfigDict(arbitrary_types_allowed=True)
@@ -135,6 +146,13 @@ class Memory(BaseModel):
             "will store memories at '/crew/research/<inferred_scope>'."
         ),
     )
+    memory_prompt: MemoryPromptConfig | None = Field(
+        default=None,
+        description=(
+            "Optional prompt strings for save, query, extract, and consolidation steps. "
+            "See MemoryPromptConfig; unset fields use translations/en.json defaults."
+        ),
+    )
 
     _config: MemoryConfig = PrivateAttr()
     _llm_instance: BaseLLM | None = PrivateAttr(default=None)
@@ -181,6 +199,7 @@ def __deepcopy__(self, memo: dict[int, Any] | None = None) -> Memory:
     def model_post_init(self, __context: Any) -> None:
         """Initialize runtime state from field values."""
         self._config = MemoryConfig(
+            memory_prompt=self.memory_prompt,
             recency_weight=self.recency_weight,
             semantic_weight=self.semantic_weight,
             importance_weight=self.importance_weight,
@@ -638,7 +657,9 @@ def extract_memories(self, content: str) -> list[str]:
         Returns:
             List of short, self-contained memory statements.
         """
-        return extract_memories_from_content(content, self._llm)
+        return extract_memories_from_content(
+            content, self._llm, self._config.memory_prompt
+        )
 
     def recall(
         self,
diff --git a/lib/crewai/src/crewai/telemetry/telemetry.py b/lib/crewai/src/crewai/telemetry/telemetry.py
index 7809c5b4ce..c93a06ff2a 100644
--- a/lib/crewai/src/crewai/telemetry/telemetry.py
+++ b/lib/crewai/src/crewai/telemetry/telemetry.py
@@ -51,6 +51,7 @@
     add_crew_and_task_attributes,
     add_crew_attributes,
     close_span,
+    crew_memory_span_attribute_value,
 )
 from crewai.utilities.logger_utils import suppress_warnings
 from crewai.utilities.string_utils import sanitize_tool_name
@@ -280,7 +281,11 @@ def _operation() -> None:
             self._add_attribute(span, "python_version", platform.python_version())
             add_crew_attributes(span, crew, self._add_attribute)
             self._add_attribute(span, "crew_process", crew.process)
-            self._add_attribute(span, "crew_memory", crew.memory)
+            self._add_attribute(
+                span,
+                "crew_memory",
+                crew_memory_span_attribute_value(crew.memory),
+            )
             self._add_attribute(span, "crew_number_of_tasks", len(crew.tasks))
             self._add_attribute(span, "crew_number_of_agents", len(crew.agents))
 
diff --git a/lib/crewai/src/crewai/telemetry/utils.py b/lib/crewai/src/crewai/telemetry/utils.py
index c6b649a302..0e818d94d0 100644
--- a/lib/crewai/src/crewai/telemetry/utils.py
+++ b/lib/crewai/src/crewai/telemetry/utils.py
@@ -16,6 +16,19 @@
     from crewai.task import Task
 
 
+def crew_memory_span_attribute_value(memory: Any) -> bool | str:
+    """Serialize ``Crew.memory`` for OpenTelemetry span attributes.
+
+    OTLP only allows bool, str, bytes, int, float, and homogeneous sequences
+    of those types — not arbitrary objects like :class:`~crewai.memory.unified_memory.Memory`.
+    """
+    if memory is None or memory is False:
+        return False
+    if memory is True:
+        return True
+    return type(memory).__name__
+
+
 def add_agent_fingerprint_to_span(
     span: Span, agent: Any, add_attribute_fn: Callable[[Span, str, Any], None]
 ) -> None:
diff --git a/lib/crewai/tests/memory/test_unified_memory.py b/lib/crewai/tests/memory/test_unified_memory.py
index f36bf0c2b9..7c7ef43453 100644
--- a/lib/crewai/tests/memory/test_unified_memory.py
+++ b/lib/crewai/tests/memory/test_unified_memory.py
@@ -664,6 +664,53 @@ def test_remember_survives_llm_failure(
     assert mem._storage.count() == 1
 
 
+# --- Per-Memory prompt config (MemoryPromptConfig) ---
+
+
+def test_memory_prompt_overrides_for_online_people_research() -> None:
+    from crewai.memory.types import MemoryPromptConfig
+
+    po = MemoryPromptConfig.for_online_people_research()
+    assert po.save_system and "search_query" in po.save_system
+    assert po.extract_memories_system and "Exa" in po.extract_memories_system
+    assert po.query_system and "recall_queries" in po.query_system
+
+
+def test_memory_prompt_overrides_save_system_used_in_analyze(tmp_path: Path) -> None:
+    from crewai.memory.analyze import analyze_for_save
+    from crewai.memory.types import MemoryPromptConfig
+    from crewai.memory.unified_memory import Memory
+
+    custom_system = "CUSTOM_SAVE_SYSTEM_OVERRIDE"
+    llm = MagicMock()
+    llm.supports_function_calling.return_value = False
+    llm.call.return_value = (
+        '{"suggested_scope": "/", "categories": [], "importance": 0.5, '
+        '"extracted_metadata": {"entities": [], "dates": [], "topics": []}}'
+    )
+
+    mem = Memory(
+        storage=str(tmp_path / "ov_db"),
+        embedder=MagicMock(),
+        llm=llm,
+        memory_prompt=MemoryPromptConfig(save_system=custom_system),
+    )
+    assert mem._config.memory_prompt is not None
+    assert mem._config.memory_prompt.save_system == custom_system
+
+    analyze_for_save(
+        "hello",
+        existing_scopes=["/"],
+        existing_categories=[],
+        llm=llm,
+        memory_prompt=mem._config.memory_prompt,
+    )
+    call_args = llm.call.call_args
+    messages = call_args[0][0]
+    assert messages[0]["role"] == "system"
+    assert messages[0]["content"] == custom_system
+
+
 # --- Agent.kickoff() memory integration ---
 
 
diff --git a/lib/crewai/tests/telemetry/test_telemetry.py b/lib/crewai/tests/telemetry/test_telemetry.py
index d0564982d2..c72f0fd443 100644
--- a/lib/crewai/tests/telemetry/test_telemetry.py
+++ b/lib/crewai/tests/telemetry/test_telemetry.py
@@ -3,8 +3,9 @@
 from unittest.mock import patch
 
 import pytest
-from crewai import Agent, Crew, Task
+from crewai import Agent, Crew, Memory, Task
 from crewai.telemetry import Telemetry
+from crewai.telemetry.utils import crew_memory_span_attribute_value
 from opentelemetry import trace
 
 
@@ -159,3 +160,20 @@ def init_in_thread():
     mock_holder["logger"].debug.assert_any_call(
         "Skipping signal handler registration: not running in main thread"
     )
+
+
+@pytest.mark.parametrize(
+    ("memory", "expected"),
+    [
+        (False, False),
+        (None, False),
+        (True, True),
+    ],
+)
+def test_crew_memory_span_attribute_value_primitives(memory, expected):
+    assert crew_memory_span_attribute_value(memory) is expected
+
+
+def test_crew_memory_span_attribute_value_memory_instance():
+    """Custom Memory instances must become a primitive string for OTLP."""
+    assert crew_memory_span_attribute_value(Memory()) == "Memory"

From 88cbf6bd1ae9223686a471da1502925c4b75129b Mon Sep 17 00:00:00 2001
From: lorenzejay <lorenzejaytech@gmail.com>
Date: Tue, 7 Apr 2026 17:53:36 -0700
Subject: [PATCH 2/3] refactor: update MemoryPromptConfig to allow custom
 prompt strings

* Removed the static method for online people research and replaced it with a constructor for MemoryPromptConfig that accepts custom strings for save, extract, and query systems.
* Updated the corresponding test to validate the new configuration approach, ensuring flexibility in memory prompt handling.
---
 lib/crewai/src/crewai/memory/types.py         | 98 -------------------
 .../tests/memory/test_unified_memory.py       | 15 ++-
 2 files changed, 10 insertions(+), 103 deletions(-)

diff --git a/lib/crewai/src/crewai/memory/types.py b/lib/crewai/src/crewai/memory/types.py
index 174ca2bf11..469fd5bdca 100644
--- a/lib/crewai/src/crewai/memory/types.py
+++ b/lib/crewai/src/crewai/memory/types.py
@@ -153,104 +153,6 @@ class MemoryPromptConfig(BaseModel):
     consolidation_system: str | None = None
     consolidation_user: str | None = None
 
-    @classmethod
-    def for_online_people_research(cls) -> MemoryPromptConfig:
-        """Prompt profile for open-web people research (e.g. Exa).
-
-        Biases memory toward **search provenance** (queries, domains, URLs, dead
-        ends, strong sources) and away from storing fragile biographical detail
-        or speculation as high-confidence fact. Pair with ``EXASearchTool`` and
-        explicit scopes/categories in ``remember()`` when you need determinism.
-        """
-        return cls(
-            save_system=(
-                "You analyze content for a hierarchical memory used during "
-                "online people research.\n"
-                "Prioritize durable *research-process* signals over raw biographical claims:\n"
-                "- suggested_scope: use paths like /subjects/<slug>, /searches/queries, "
-                "/sources/domains when the content is about where or how you searched.\n"
-                "- categories: include tags such as search_query, result_domain, "
-                "source_url, exa_search, dead_end, preferred_source, methodology "
-                "when applicable. Use subject_profile only for well-sourced factual "
-                "snippets tied to a named URL or document - not for gossip or guesses.\n"
-                "- importance: higher (0.7-1.0) for reusable search lessons "
-                "(e.g. 'site X returned 404', 'LinkedIn useful for role Y', "
-                "'query phrasing Z worked'). Lower (0.2-0.5) for one-off personal "
-                "minutiae or unverified claims.\n"
-                "- extracted_metadata.entities: prefer domains, publication names, "
-                "and search keywords; only list people when the text explicitly "
-                "anchors a fact to a source.\n"
-                "Given the content and existing scopes and categories, output:\n"
-                "1. suggested_scope\n2. categories\n3. importance\n4. extracted_metadata"
-            ),
-            save_user=(
-                "Content to store:\n{content}\n\n"
-                "Existing scopes: {existing_scopes}\n"
-                "Existing categories: {existing_categories}\n\n"
-                "Return the analysis as structured output."
-            ),
-            query_system=(
-                "You analyze a query against memory for people-research crews.\n"
-                "Favor recalling *how and where* the team already searched:\n"
-                "- keywords: include domains, site names, query stems, and subject "
-                "handles - not only the person's name.\n"
-                "- suggested_scopes: prefer /searches, /sources, /subjects as relevant.\n"
-                "- recall_queries: produce phrases that would match memories about "
-                "Exa queries, domains tried, and conclusions about source quality.\n"
-                "- complexity: 'complex' if the question asks to synthesize many sources; "
-                "else 'simple'.\n"
-                "- time_filter: ISO date if the query implies a time window; else null.\n"
-                "Given the query and available scopes, output:\n"
-                "1. keywords\n2. suggested_scopes\n3. complexity\n4. recall_queries\n"
-                "5. time_filter"
-            ),
-            query_user=(
-                "Query: {query}\n\n"
-                "Available scopes: {available_scopes}\n"
-                "{scope_desc}\n\n"
-                "Return the analysis as structured output."
-            ),
-            extract_memories_system=(
-                "Extract discrete memory statements from research logs (tool outputs, "
-                "Exa results, agent notes).\n"
-                "HIGH priority - always capture when present:\n"
-                "- Exact or paraphrased search queries used and which engine/tool "
-                "(e.g. Exa).\n"
-                "- Domains and canonical URLs of pages opened; HTTP errors or empty results.\n"
-                "- Judgments about source usefulness (e.g. 'official bio on company site', "
-                "'social profile - verify elsewhere').\n"
-                "MEDIUM priority:\n"
-                "- Verifiable facts explicitly tied to a source in the text "
-                "(quote which source).\n"
-                "LOW priority / often skip:\n"
-                "- Speculation, reputation gossip, or detailed private life unless "
-                "the task explicitly requires it and a primary source is named.\n"
-                "Each memory: one clear sentence, no duplicate ideas.\n"
-                'Output JSON with a single key "memories" whose value is a list of strings.'
-            ),
-            extract_memories_user=(
-                "Content:\n{content}\n\n"
-                "Extract memory statements as described. Return structured output."
-            ),
-            consolidation_system=(
-                "You consolidate new research content with similar existing memories.\n"
-                "Prefer keeping separate memories when they differ by URL, domain, or "
-                "search query - even if about the same person.\n"
-                "Merge or update when the new text clearly refines the same source "
-                "or same search finding.\n"
-                "Delete only when a memory is clearly wrong and superseded.\n"
-                "Be conservative: prefer 'keep' when unsure.\n"
-                "For each existing memory: keep | update | delete. "
-                "Set insert_new when the new content adds a distinct query, URL, or "
-                "source judgment."
-            ),
-            consolidation_user=(
-                "New content to consider storing:\n{new_content}\n\n"
-                "Existing similar memories:\n{records_summary}\n\n"
-                "Return the consolidation plan as structured output."
-            ),
-        )
-
 
 class MemoryConfig(BaseModel):
     """Internal configuration for memory scoring, consolidation, and recall behavior.
diff --git a/lib/crewai/tests/memory/test_unified_memory.py b/lib/crewai/tests/memory/test_unified_memory.py
index dea7760b1b..eec83f77d3 100644
--- a/lib/crewai/tests/memory/test_unified_memory.py
+++ b/lib/crewai/tests/memory/test_unified_memory.py
@@ -653,13 +653,18 @@ def test_remember_survives_llm_failure(
 # --- Per-Memory prompt config (MemoryPromptConfig) ---
 
 
-def test_memory_prompt_overrides_for_online_people_research() -> None:
+def test_memory_prompt_config_custom_strings() -> None:
+    """Library stays domain-agnostic; apps pass their own MemoryPromptConfig."""
     from crewai.memory.types import MemoryPromptConfig
 
-    po = MemoryPromptConfig.for_online_people_research()
-    assert po.save_system and "search_query" in po.save_system
-    assert po.extract_memories_system and "Exa" in po.extract_memories_system
-    assert po.query_system and "recall_queries" in po.query_system
+    po = MemoryPromptConfig(
+        save_system="Prefer categories: search_query, exa_search, result_domain.",
+        extract_memories_system="Record Exa queries and canonical URLs first.",
+        query_system="Distill recall_queries toward domains and past queries.",
+    )
+    assert "search_query" in (po.save_system or "")
+    assert "Exa" in (po.extract_memories_system or "")
+    assert "recall_queries" in (po.query_system or "")
 
 
 def test_memory_prompt_overrides_save_system_used_in_analyze(tmp_path: Path) -> None:

From 77dcf265b67866524b17a7f5a0d3542b907f90be Mon Sep 17 00:00:00 2001
From: lorenzejay <lorenzejaytech@gmail.com>
Date: Wed, 22 Apr 2026 16:13:40 -0700
Subject: [PATCH 3/3] added docs

---
 docs/ar/concepts/memory.mdx    | 37 ++++++++++++++++++++++++++++++++++
 docs/en/concepts/memory.mdx    | 37 ++++++++++++++++++++++++++++++++++
 docs/ko/concepts/memory.mdx    | 37 ++++++++++++++++++++++++++++++++++
 docs/pt-BR/concepts/memory.mdx | 37 ++++++++++++++++++++++++++++++++++
 4 files changed, 148 insertions(+)

diff --git a/docs/ar/concepts/memory.mdx b/docs/ar/concepts/memory.mdx
index 541f2967a7..d3a93c3c74 100644
--- a/docs/ar/concepts/memory.mdx
+++ b/docs/ar/concepts/memory.mdx
@@ -157,6 +157,43 @@ class ResearchFlow(Flow):
 
 انظر [وثائق التدفقات](/concepts/flows) لمزيد من المعلومات حول الذاكرة في التدفقات.
 
+## تخصيص مطالبات الذاكرة (`MemoryPromptConfig`)
+
+يمكنك استبدال تعليمات نموذج اللغة في كل خطوة من تحليل الذاكرة (نفس فكرة ضبط مطالبات التخطيط). مرّر كائن `MemoryPromptConfig` كوسيط `memory_prompt` إلى `Memory`. عيّن الحقول التي تحتاجها فقط؛ تبقى الخطوات الأخرى على القيم الافتراضية المضمّنة في `translations/en.json` تحت المفتاح `memory` (أسماء الحقول تطابق مفاتيح JSON).
+
+```python
+from crewai import Memory, MemoryPromptConfig
+
+memory = Memory(
+    llm="gpt-4o-mini",
+    memory_prompt=MemoryPromptConfig(
+        save_system="...",   # اختياري
+        query_user="...",    # اختياري
+    ),
+)
+```
+
+يمكنك أيضًا تمرير `memory_prompt` إلى دوال مساعدة في `crewai.memory.analyze` (مثل `extract_memories_from_content`) عند استدعائها مباشرة.
+
+### تأثير كل زوج من المطالبات
+
+| الحقول | متى يعمل | ماذا يؤثر |
+| --- | --- | --- |
+| `save_system` / `save_user` | عند الحفظ (`analyze_for_save`) | `suggested_scope` و`categories` و`importance` و`extracted_metadata` المستنتجة قبل التخزين والتضمين. |
+| `query_system` / `query_user` | عند تحليل استعلام الاسترجاع (`analyze_query`) | `keywords` و`suggested_scopes` و`complexity` و`recall_queries` و`time_filter`، ما يوجّه البحث المتجهي وعمق الاسترجاع. |
+| `extract_memories_system` / `extract_memories_user` | `extract_memories_from_content` / `Memory.extract_memories` | كيفية تقسيم النص الخام إلى جمل ذاكرة منفصلة (لا يزال التخزين عبر `remember()`). |
+| `consolidation_system` / `consolidation_user` | عندما يكون المحتوى الجديد قريبًا دلاليًا من سجلات موجودة (`analyze_for_consolidation`) | الإبقاء على الصفوف أو تحديثها أو حذفها، وما إذا كان يُدرج المحتوى الجديد كذاكرة مستقلة. |
+
+### العناصر النائبة (placeholders)
+
+سلاسل **النظام (system)** تُرسل كما هي. سلاسل **المستخدم (user)** تُملأ بـ `str.format` في بايثون. يجب أن تتضمن قوالب المستخدم المخصصة نفس أسماء العناصر النائبة الافتراضية وإلا يفشل التنسيق.
+
+| حقل المستخدم | عناصر نائبة مطلوبة |
+| --- | --- |
+| `save_user` | `{content}`، `{existing_scopes}`، `{existing_categories}` |
+| `query_user` | `{query}`، `{available_scopes}`، `{scope_desc}` |
+| `extract_memories_user` | `{content}` |
+| `consolidation_user` | `{new_content}`، `{records_summary}` |
 
 ## النطاقات الهرمية
 
diff --git a/docs/en/concepts/memory.mdx b/docs/en/concepts/memory.mdx
index 954d5efe6e..04a1192c54 100644
--- a/docs/en/concepts/memory.mdx
+++ b/docs/en/concepts/memory.mdx
@@ -157,6 +157,43 @@ class ResearchFlow(Flow):
 
 See the [Flows documentation](/concepts/flows) for more on memory in Flows.
 
+## Customizing memory prompts (`MemoryPromptConfig`)
+
+Override the LLM instructions used at each memory analysis step (same idea as tuning planning prompts). Pass a `MemoryPromptConfig` as `memory_prompt` on `Memory`. Only set the fields you need; every other step keeps the bundled defaults from the library’s `translations/en.json` under the `memory` key (field names match those JSON keys).
+
+```python
+from crewai import Memory, MemoryPromptConfig
+
+memory = Memory(
+    llm="gpt-4o-mini",
+    memory_prompt=MemoryPromptConfig(
+        save_system="...",   # optional
+        query_user="...",    # optional
+    ),
+)
+```
+
+You can also pass `memory_prompt` into helpers in `crewai.memory.analyze` (for example `extract_memories_from_content`) when you call them directly.
+
+### What each prompt pair affects
+
+| Fields | When it runs | What it influences |
+| --- | --- | --- |
+| `save_system` / `save_user` | Saving (`analyze_for_save`) | Inferred `suggested_scope`, `categories`, `importance`, and `extracted_metadata` before storage and embedding. |
+| `query_system` / `query_user` | Recall query analysis (`analyze_query`) | `keywords`, `suggested_scopes`, `complexity`, `recall_queries`, and `time_filter`, which steer vector search and how deep recall goes. |
+| `extract_memories_system` / `extract_memories_user` | `extract_memories_from_content` / `Memory.extract_memories` | How raw text is split into discrete memory strings (persistence is still via `remember()`). |
+| `consolidation_system` / `consolidation_user` | When new content is similar to existing records (`analyze_for_consolidation`) | Whether to keep, update, or delete existing rows and whether to insert the new content as its own memory. |
+
+### Placeholders
+
+**System** strings are sent as-is. **User** strings are filled with Python’s `str.format`. Custom user templates must include the same placeholder names as the defaults or formatting will raise.
+
+| User field | Required placeholders |
+| --- | --- |
+| `save_user` | `{content}`, `{existing_scopes}`, `{existing_categories}` |
+| `query_user` | `{query}`, `{available_scopes}`, `{scope_desc}` |
+| `extract_memories_user` | `{content}` |
+| `consolidation_user` | `{new_content}`, `{records_summary}` |
 
 ## Hierarchical Scopes
 
diff --git a/docs/ko/concepts/memory.mdx b/docs/ko/concepts/memory.mdx
index ea4463eea7..37578be28b 100644
--- a/docs/ko/concepts/memory.mdx
+++ b/docs/ko/concepts/memory.mdx
@@ -157,6 +157,43 @@ class ResearchFlow(Flow):
 
 Flow에서의 메모리에 대한 자세한 내용은 [Flows 문서](/concepts/flows)를 참조하세요.
 
+## 메모리 프롬프트 사용자 지정 (`MemoryPromptConfig`)
+
+메모리 분석 단계마다 사용되는 LLM 지시문을 덮어쓸 수 있습니다(플래닝 프롬프트를 조정하는 것과 같은 개념). `Memory`의 `memory_prompt`에 `MemoryPromptConfig`를 넘깁니다. 필요한 필드만 설정하면 되고, 나머지 단계는 라이브러리 번들 기본값(`translations/en.json`의 `memory` 키; 필드 이름이 해당 JSON 키와 일치)을 그대로 씁니다.
+
+```python
+from crewai import Memory, MemoryPromptConfig
+
+memory = Memory(
+    llm="gpt-4o-mini",
+    memory_prompt=MemoryPromptConfig(
+        save_system="...",   # 선택
+        query_user="...",    # 선택
+    ),
+)
+```
+
+`crewai.memory.analyze`의 헬퍼(예: `extract_memories_from_content`)를 직접 호출할 때도 `memory_prompt`를 넘길 수 있습니다.
+
+### 프롬프트 쌍별 역할
+
+| 필드 | 실행 시점 | 영향 |
+| --- | --- | --- |
+| `save_system` / `save_user` | 저장 시 (`analyze_for_save`) | 저장·임베딩 전에 추론되는 `suggested_scope`, `categories`, `importance`, `extracted_metadata`. |
+| `query_system` / `query_user` | 리콜 시 쿼리 분석 (`analyze_query`) | `keywords`, `suggested_scopes`, `complexity`, `recall_queries`, `time_filter` — 벡터 검색과 리콜 탐색 깊이에 영향. |
+| `extract_memories_system` / `extract_memories_user` | `extract_memories_from_content` / `Memory.extract_memories` | 긴 텍스트를 개별 메모리 문자열로 나누는 방식(저장은 여전히 `remember()`). |
+| `consolidation_system` / `consolidation_user` | 신규 콘텐츠가 기존 레코드와 유사할 때 (`analyze_for_consolidation`) | 기존 행 유지·갱신·삭제 및 신규 콘텐츠를 별도 메모리로 넣을지 여부. |
+
+### 플레이스홀더
+
+**system** 문자열은 그대로 전송됩니다. **user** 문자열은 Python `str.format`으로 채워집니다. 사용자 정의 user 템플릿에는 기본값과 동일한 플레이스홀더 이름이 포함되어야 하며, 그렇지 않으면 포맷 단계에서 오류가 납니다.
+
+| User 필드 | 필수 플레이스홀더 |
+| --- | --- |
+| `save_user` | `{content}`, `{existing_scopes}`, `{existing_categories}` |
+| `query_user` | `{query}`, `{available_scopes}`, `{scope_desc}` |
+| `extract_memories_user` | `{content}` |
+| `consolidation_user` | `{new_content}`, `{records_summary}` |
 
 ## 계층적 범위(Scopes)
 
diff --git a/docs/pt-BR/concepts/memory.mdx b/docs/pt-BR/concepts/memory.mdx
index 3931ed6ab5..98b624cd67 100644
--- a/docs/pt-BR/concepts/memory.mdx
+++ b/docs/pt-BR/concepts/memory.mdx
@@ -157,6 +157,43 @@ class ResearchFlow(Flow):
 
 Veja a [documentação de Flows](/concepts/flows) para mais informações sobre memória em Flows.
 
+## Personalizando prompts de memória (`MemoryPromptConfig`)
+
+Substitua as instruções do LLM usadas em cada etapa de análise de memória (mesma ideia que ajustar prompts de planejamento). Passe um `MemoryPromptConfig` como `memory_prompt` em `Memory`. Defina apenas os campos necessários; nas demais etapas permanecem os padrões embutidos do `translations/en.json` da biblioteca, na chave `memory` (os nomes dos campos correspondem às chaves JSON).
+
+```python
+from crewai import Memory, MemoryPromptConfig
+
+memory = Memory(
+    llm="gpt-4o-mini",
+    memory_prompt=MemoryPromptConfig(
+        save_system="...",   # opcional
+        query_user="...",    # opcional
+    ),
+)
+```
+
+Você também pode passar `memory_prompt` para funções auxiliares em `crewai.memory.analyze` (por exemplo `extract_memories_from_content`) quando chamá-las diretamente.
+
+### O que cada par de prompts afeta
+
+| Campos | Quando roda | O que influencia |
+| --- | --- | --- |
+| `save_system` / `save_user` | Ao salvar (`analyze_for_save`) | `suggested_scope`, `categories`, `importance` e `extracted_metadata` inferidos antes do armazenamento e do embedding. |
+| `query_system` / `query_user` | Análise da consulta no recall (`analyze_query`) | `keywords`, `suggested_scopes`, `complexity`, `recall_queries` e `time_filter`, que orientam a busca vetorial e a profundidade do recall. |
+| `extract_memories_system` / `extract_memories_user` | `extract_memories_from_content` / `Memory.extract_memories` | Como o texto bruto é dividido em memórias atômicas (a persistência continua sendo via `remember()`). |
+| `consolidation_system` / `consolidation_user` | Quando o novo conteúdo é semelhante a registros existentes (`analyze_for_consolidation`) | Manter, atualizar ou excluir linhas existentes e se o novo conteúdo entra como memória própria. |
+
+### Placeholders
+
+Strings de **system** são enviadas como estão. Strings de **user** são preenchidas com `str.format` do Python. Templates de user personalizados devem incluir os mesmos nomes de placeholder dos padrões; caso contrário, a formatação falha.
+
+| Campo user | Placeholders obrigatórios |
+| --- | --- |
+| `save_user` | `{content}`, `{existing_scopes}`, `{existing_categories}` |
+| `query_user` | `{query}`, `{available_scopes}`, `{scope_desc}` |
+| `extract_memories_user` | `{content}` |
+| `consolidation_user` | `{new_content}`, `{records_summary}` |
 
 ## Escopos Hierárquicos