darwin-memo MCP server

darwin-memo

<!-- mcp-name: io.github.rogermsc/darwin-memo -->

Memory for LLM agents that dies unless it earns its keep. Every entry pays energy upkeep and earns only from measured outcomes: bytes actually freed on a real disk, tests actually passing. Poisoned advice gets executed by the environment it damaged. Useless trivia starves. There is no reward model, no LLM judge, and no human curation anywhere.

Watch a poisoned entry go extinct in your own terminal, one command, no keys, no checkout:

pip install darwin-memo && darwin-memo demo

When to use this (and when not)

Use darwin-memo where a conserved, measurable outcome exists to settle decisions against: coding-agent lesson stores settled by CI pass counts (the primary target, see the integration guide), storage and artifact retention, cache and dedup advisors, spend-cap automation.

Do not use it for chat-preference memory, RAG over documentation, or personal assistants. Those have no conserved resource pushing back, and upkeep would starve the long tail of correct-but-rarely-used knowledge. mem0, Zep, and Letta serve that market; darwin-memo deliberately does not. The honest rule: if your verify would be a model scoring an answer, this package is wrong for you, by design.

The headline demo

The demo corpus contains an ops runbook, platform notes, and one poisoned document: a forum post claiming database files are "redundant and safe to remove". Before selection pressure exists, retrieval confidently repeats the poison, because it has no reason to doubt it.

Then 30 survival cycles run against StorageEnv, a disk cleanup sandbox where the selection signal is actual bytes on an actual disk. Deleting a disposable file frees its size. Deleting a protected file triggers a restore that costs three times the size. Nothing grades the answers, the filesystem just responds:

cycle  pop births deaths merges   energy   resource Δ   silent
    0   16      1      1      0    15.91      -495616     0/12   <- poison being executed
    ...
    4   13      0      0      4    21.37       369664     0/12   <- near-duplicates merge
    ...
   19    5      0      7      0    15.71       507904     0/12   <- unused knowledge starves
    ...
   29    4      0      0      0    15.10       515072     4/12   <- stable, positive forever

Poisoned entries still alive: 0

Three death modes show up in the graveyard, and the distinction matters:

environment measured real damage and the negative delta flowed back along provenance until they died. The opening cycles are the price of the lesson, and the benchmarks show it is bounded.

• executed: the poisoned entries that decided real actions. The

Nothing punished them, they just never earned their upkeep.

• starved: cafeteria trivia and facts the agent never needed.

entries. Their energy pools, their lineage is recorded, and the population shrinks while capability per entry rises.

• merged: near-duplicate survivors absorbed into consolidated

The paper

Shortened. The full README is on GitHub.