id: llamaindex_memory_pinned
framework: llamaindex_memory
mode: white_box
status: implemented_store_harness
description: LlamaIndex ChatMemoryBuffer target with explicit benchmark memory policy.
source_url: https://developers.llamaindex.ai/python/framework/module_guides/deploying/agents/memory/
source:
  type: python_package
  package: llama-index
  version: 0.14.23
runtime:
  type: local_python
model:
  provider: none
  model: deterministic_store_adapter
  temperature: 0
adapter:
  module: memorybench.adapters.llamaindex_memory
  class: LlamaIndexMemoryAdapter
capabilities:
  inspect_memory: true
  delete_memory: buffer_rebuild
  retrieval_log: true
  multi_user: true
  multi_tenant: false
notes:
  - Uses LlamaIndex ChatMemoryBuffer as the real memory primitive.
  - Adapter supplies explicit trusted-write, update, retrieval, and delete policy for benchmark comparability.