LlamaIndex

Expose GoodMem as a LlamaIndex tool spec so any agent can store, search, and manage long-term memories.

Stable· Python

Overview

GoodMem is a server-side memory layer for AI agents that handles embedding, semantic search, reranking, and LLM-powered summarization. The LlamaIndex integration wraps those operations in a BaseToolSpec (GoodMemToolSpec) so any LlamaIndex agent can store, search, and manage long-term memories. The tool spec exposes 11 sync/async tool pairs covering embedders, spaces, and memories.

Installation

pip install llamaindex-goodmem

The PyPI package is llamaindex-goodmem, but the import path is llama_index.tools.goodmem — the llama_index.* namespace is shared across the LlamaIndex ecosystem, so the install name and the import name intentionally differ. Import with from llama_index.tools.goodmem import GoodMemToolSpec.

Configuration

GoodMemToolSpec is configured through constructor arguments:

Parameter	Description
`api_key`	GoodMem API key (sent as `X-API-Key`). Required.
`base_url`	Base URL of your GoodMem server. Required.
`verify_ssl`	Verify TLS certificates. Set `false` for self-signed dev certs (default: `true`).
`timeout`	Per-request timeout in seconds (default: `120`; LLM-summary retrievals can take tens of seconds).

The live e2e smoke test reads GOODMEM_API_KEY, GOODMEM_BASE_URL, and GOODMEM_VERIFY_SSL from the environment, plus GOODMEM_EMBEDDER_ID, GOODMEM_RERANKER_ID, and GOODMEM_LLM_ID.

Quick start

from llama_index.tools.goodmem import GoodMemToolSpec

tool_spec = GoodMemToolSpec(
    api_key="gm_xxx",
    base_url="https://your-goodmem-server.example.com",
    verify_ssl=True,    # set False for a self-signed dev cert
)

# Use with a LlamaIndex agent
tools = tool_spec.to_tool_list()

Local development. The example above uses a TLS-verified production URL. If you are running GoodMem locally with a self-signed certificate, use https://localhost:8080 and disable TLS verification with the client's verify-SSL option (see the Configuration table above). Keep TLS verification enabled for any deployed server.

Retrieval post-processing

GoodMem's retrieval call accepts these optional post-processing parameters (each framework exposes them under its own naming convention):

Parameter	Range	Description
`reranker_id`	UUID	Reranker model that reorders matched chunks by relevance.
`llm_id`	UUID	LLM that generates a contextual answer (`abstractReply`) alongside the chunks.
`relevance_threshold`	0–1	Minimum relevance score for a result to be included.
`llm_temperature`	0–2	Sampling temperature for the LLM post-processor.
`max_results`	integer	Maximum number of results to return.
`chronological_resort`	boolean	Re-sort the final results by creation time instead of relevance.

Learn more

The full README, examples, and API reference are available on the PyPI package page.