memory-boost

memory-boost

Memory for coding agents that reviews itself.

_{drift, mine and lessons on the bundled example data · everything runs locally}

Try it · What it answers · Connect your agent · How it works · vs. alternatives

Your agents write everything down. Nobody reads it back.

Every memory tool for coding agents helps them remember more. None of them mention that half of what they remember expired in July. memory-boost keeps a plain-markdown memory your agents read and write over MCP, and then does the unglamorous part: it audits that memory and your own sessions, and tells you what drifted. Politely, with numbers.

Think of it as a code review for what your agents remember and a fitness tracker for how you work with them. Mostly the fitness tracker: mine told me my most-failed command was cd. Sixty-nine times.

What it answers

You wonder…	memory-boost answers
Which decisions quietly expired while we were busy shipping?	drift: review_by 2026-07-08, 78 days ago
Which "active" projects actually died months ago?	drift: last session 161 days ago; still listed as live
We fixed this exact bug in another repo. How?	lessons: the fix follows you to every project with the same stack
Why does my agent keep re-reading the same file?	mine: one file, read 18 times
Is my agent spawning 20 subagents behind my back?	mine: yes, and in which session
Where was I before /compact ate my context?	The session-start hook puts your checkpoint back

Try it in 30 seconds

No install needed; the wiki lands in ~/.local/share/memory-boost:

uvx memory-boost==0.2.0 init --example   # sample wiki with five kinds of rot + four synthetic transcripts
uvx memory-boost==0.2.0 drift            # what in that memory is stale, orphaned or never written down
uvx memory-boost==0.2.0 mine             # now on your real ~/.claude/projects: aggregates only

To keep it in Claude Code, install the plugin. It brings the MCP server, the session-start brief and the skills in one step; it only needs uv on your PATH.

/plugin marketplace add arnaugonzalez/memory-boost
/plugin install memory-boost@memory-boost

Other agents, or the CLI on its own: uv tool install memory-boost==0.2.0 (or pipx install memory-boost==0.2.0). Or let your agent do it: see Install from your agent.

Install from your agent

Paste this into Claude Code, Codex, Cline or Cursor:

Install memory-boost 0.2.0 by following https://github.com/arnaugonzalez/memory-boost/blob/main/llms-install.md : install the CLI with uv tool install memory-boost==0.2.0, run memory-boost init, register memory-boost serve as a stdio MCP server named memory, check that it answers with nine memory_* tools, then run memory-boost drift and show me the output. Ask me before writing outside the memory home (~/.local/share/memory-boost) or editing any global agent config.

The agent-oriented steps (stdio and serve --http, per harness, with checks) are in llms-install.md.

On a real machine

Five weeks of the author's own daily use: 29 decisions, 24 projects, 657 Claude Code transcripts. Output as produced; only project and file names are replaced.

$ memory-boost drift
# Drift report — 2026-09-24
29 decisions · 24 projects · 204 log entries · 33 open checkpoints · **20 findings**

## decision-expired (9)
- deploy-via-bastion — review_by 2026-07-08 (78 days ago); revisit or mark historic
- bot-guardrails — review_by 2026-07-08 (78 days ago); revisit or mark historic
  …
## project-no-decisions (2)
- thesis-solver — 20 sessions logged, 0 decisions tagged 'thesis-solver': the learning lives only in transcripts
## project-silent (5)
- mood-api — last session 2026-04-16 (161 days ago); still listed as live
  …

$ memory-boost mine
# Session digest — 2026-09-24
sessions      80 with ≥20 turns
tools         Bash 5956 · Edit 860 · Read 692 · Write 493 · Agent 203 · ToolSearch 138
bash          5956 calls, 159 failed (2.7%); most failed: cd (69), sleep (12), ssh (9), pkill (8), ls (7)
bash/edit     4.4x; extreme: crm-wt-f11 107.5x, sandbox 70.0x, sandbox 54.0x
reads         692 calls, 131 re-reads of a file already read (19%); top: …/apply_email.py ×18
discipline    62 long sessions (≥60 turns): 47 never checkpointed, 29 never saved
subagents     203 spawned in 44 sessions; max in one session: 20

Every line is a habit to change, a decision to revisit, or a rule worth writing down. None of it needed a new tool: the data was already on disk. It needed a reader.

What it is

• A wiki you own. projects/*.md, decisions/*.md, concepts/*.md, log.md. Plain markdown; edit by hand, keep it in git. SQLite FTS5 index as a disposable cache. Nine MCP tools for any harness: memory_brief, memory_recall, memory_page, memory_save, memory_checkpoint, memory_resume, memory_drift, memory_lesson, memory_shelf. memory_shelf (CLI: memory-boost shelf "idea") appends a dated one-line idea to concepts/idea-shelf.md without the agent reading the page back.
• Freshness as data. Decisions carry status: and review_by:. Search results are tagged current / unclassified / review / historic / superseded, current first.
• Session checkpoints that survive /compact, with one line per other active session on the same project so parallel agents don't collide.
• drift reads the wiki and reports what expired, what went silent, what was never written down and what was left half-done. The current project's top findings go into every session brief.
• mine reads your local Claude Code transcripts and reports tool usage, failed commands, re-reads, subagent fan-out and checkpoint discipline. Aggregates only: no prompt, command or file content is stored or printed, so the digest is safe to paste anywhere.
• lessons: concepts/ pages with applies_when: [tags]. A project page that lists matching tags: gets those lessons in its brief, so what worked in one repo reaches the next. The /retro skill writes them at session end.

Connect your harness

MCP server (all harnesses): command memory-boost, args ["serve"].

To share one server between several sessions, run it over streamable HTTP (stateless, JSON responses) and point each harness at http://127.0.0.1:8765/mcp:

memory-boost serve --http                      # --host 127.0.0.1 --port 8765 are the defaults
claude mcp add --transport http memory -s user http://127.0.0.1:8765/mcp

It binds to loopback by default and has no authentication: keep it there.

Harness	MCP registration	Context at session start
Claude Code (plugin)	/plugin install memory-boost@memory-boost	Included: hook, MCP server, skills
Claude Code (manual)	claude mcp add memory -s user -- memory-boost serve	Hook below (startup, resume, compact); mine reads its transcripts
Codex CLI	codex mcp add memory -- memory-boost serve	AGENTS.md: "call memory_brief first"
Cursor / VS Code / Gemini CLI	stdio entry in their mcp.json / settings.json	Same line in rules (hook adapters: roadmap)

Manual Claude Code hook (~/.claude/settings.json), not needed with the plugin:

{ "hooks": { "SessionStart": [ { "hooks": [ { "type": "command", "command": "memory-boost hook session-start" } ] } ] } }

On startup/resume it injects ~1k tokens: catalog, the project's context and last sessions, its drift, lessons from other projects, your open checkpoint. On compact only this session's checkpoint. It never blocks session start: on any error it prints nothing.

Skills: /memory-boost (router: when to run what), /save-session, /restore-session, /retro. The plugin installs them as /memory-boost:retro and so on; without it, copy skills/* into ~/.claude/skills/. The first session after installing the plugin takes a few extra seconds while uv fetches memory-boost==0.2.0 and its dependencies from PyPI (the only network access the plugin makes); later sessions start in about 0.2 s.

CLI contract for agents: data on stdout, errors on stderr with a fix: line, --json on every reading command (with a version field), exit 0 = trust stdout, 1 = not found, 2 = bad input. Reports end with a next: line naming the exact command to run.

How it works

_{Source: docs/architecture.mmd}

The project is resolved from the working directory name (~/work/acme-api → projects/acme-api.md); map other names in aliases.json. Under 1,900 lines of Python, one dependency (mcp), no embeddings unless you opt into semantic recall, no daemon unless you run serve --http.

Semantic recall (optional)

Off by default: memory_recall is full-text only and the base install stays one dependency. To also match paraphrases and other languages, run a local embedding model:

pip install "memory-boost[semantic]==0.2.0"   # adds sqlite-vec
ollama pull embeddinggemma
export MEMORY_BOOST_SEMANTIC=1
memory-boost index                       # embeds the wiki once; later runs embed only changes

The first index of a large wiki on CPU can take minutes: as a rough guide, about 15 minutes for some 850 documents on a laptop without a GPU (one measurement, not a benchmark). memory-boost index and recall show embedding N/M chunks… on stderr while it runs; later runs re-embed only what changed.

Default mode when on is hybrid (full-text and vectors fused by weighted reciprocal rank); override with MEMORY_BOOST_RECALL_MODE=fts|semantic|hybrid or memory-boost recall --mode. If Ollama is down, recall falls back to full-text and logs a warning. On a private 64-query set (39 literal, 25 paraphrase) with a warm index on CPU: paraphrase nDCG@10 0.46 to 0.76, literal 0.80 to 0.85, p50 about 110 ms per query. The query set is not published.

Configuration

Variable	Default	Purpose
MEMORY_BOOST_HOME	$XDG_DATA_HOME/memory-boost	Wiki, events, checkpoints, aliases.json
MEMORY_BOOST_WIKI	$MEMORY_BOOST_HOME/wiki	Point at an existing wiki (e.g. a git repo)
MEMORY_BOOST_INDEX	$XDG_CACHE_HOME/memory-boost/index-<hash>.db	Search index
MEMORY_BOOST_CHECKPOINTS	$MEMORY_BOOST_HOME/checkpoints	Checkpoint directory
MEMORY_BOOST_TRANSCRIPTS	~/.claude/projects	Where mine looks (several dirs: a:b)
MEMORY_BOOST_TODAY	today (UTC)	Pin the date for reproducible reports

vs. alternatives

	memory-boost	basic-memory	claude-mem	agent-retro
Store	Markdown wiki you edit	Markdown	Captured observations	None (prompt skill)
Search	FTS5 (+ optional local vectors)	FTS + vectors	FTS + vectors	—
Freshness / review dates	Yes	No	No	—
Drift report over memory	Yes	No	No	No
Session analytics	Yes (aggregates, local)	No	No	LLM reads the conversation
Parallel-session awareness	Yes	No	No	—
License	MIT	AGPL-3.0	Apache-2.0	?

Want better search, more integrations and a community? Use basic-memory. Want everything captured automatically? claude-mem. Use memory-boost if you want to learn from what your agents already wrote down.

Limits / non-goals

• Full-text search by default; embeddings only with the optional [semantic] extra and a local Ollama.
• mine understands Claude Code transcripts today. Codex is on the roadmap.
• Reports are as good as the wiki: an empty wiki yields an empty drift.
• Single user, local files. Linux and macOS; Windows untested.
• The wiki is read by your agents as instructions. Only use a wiki you trust; see SECURITY.md.

License

MIT