AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Home/
Memory & Context/
claude-mem
claude-mem logo

claude-mem

Active·★ 118.0k·Apache-2.0·Updated 2026-11-18
★ Most Popular

claude-mem by thedotmack is a viral open-source persistent memory system (Apache 2.0) engineered to solve one of the most frustrating bottlenecks in AI-assisted software development: the complete context amnesia coding agents experience between fresh sessions.

Rather than dumping unpruned, token-heavy conversation transcripts into subsequent prompts—which inflates API bills and exhausts context windows—Claude-Mem deploys a lightweight companion observer agent that operates in the background. As the primary coding agent implements features and fixes bugs, Claude-Mem continuously captures concise project notes, architectural decisions, and resolved gotchas, compiling them into a local hierarchical SQLite and vector memory database.

On subsequent launches across Claude Code, Codex, Cursor, OpenCode, and 11 other harnesses, Claude-Mem injects a compact index of relevant historical decisions, dynamically retrieving granular context via MCP search tools only when triggered. Teams deploying it should note that running the background note-taking daemon incurs minor auxiliary API token expenses, and massive uncommitted branch switches outside the CLI require manual memory re-indexing.

This write-up is based on the project's README and public repository data. We have not hand-tested the tool.

#ai#ai-agents#ai-memory#anthropic#artificial-intelligence#chromadb#claude#claude-agent-sdk
$ Install
$ npx claude-mem install
↗ Visit site★ GitHub
01

Features

01Hooks feed tool activity to a local worker.
02Observer-generated observations and summaries persist locally.
03MCP search retrieves details behind compact context indexes.
02

Why choose it

+Hierarchical memory indexing injects minimal token-efficient summaries, retrieving details only on demand.
+Runs as an autonomous background companion observer, seamlessly taking notes without interrupting active workflows.
+Broad cross-agent versatility supporting Claude Code, Codex, Cursor, OpenCode, Copilot CLI, and 11 other harnesses.
+Native Model Context Protocol (MCP) search integration allows agents to query historical decisions conversationally.
03

Trade-offs

!Background memory synthesis and vector embedding generation incur modest secondary LLM API token costs.
!Requires running a persistent local background process and maintaining a local SQLite/vector storage file.
!Large manual git branch checkouts performed outside the agent CLI can lead to temporary memory desynchronization.
!Memory retrieval accuracy depends on the quality of natural language queries formulated by the host agent.
04

Maintenance check

ActiveChecked: 2026-11-12
Last commit
2026-11-12
Commits, last 90 days
100+
Releases, last 90 days
30
Latest release
v15.18.3 (2026-11-12)
Open issues
87
Recent issues with at least one comment
4/5
Commits per week, last 12 weeks
Commits per week, last 12 weeks: 29, 57, 153, 8, 30, 248, 217, 194, 222, 123, 241, 581

Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.

Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.

Most discussed open issues

  • #3606 [plan-18] Observer Response Pipeline — classify every output, never confirm a batch on failure, bound every history45 comments · 2026-08-16
  • #3607 [plan-19] Observer Subprocess Isolation — an explicit env, cwd, config, tool set and auth source for the headless generator, with loud degra25 comments · 2026-08-16
  • #3603 [plan-15] Worker Port & Liveness Authority — one verified answer to "is a worker serving this port?"24 comments · 2026-08-16
  • #3609 [plan-21] SQLite Schema Evolution & Queue State Integrity — one DDL path guarded by introspection, immutable session keys, and queue/session23 comments · 2026-08-16
  • #3602 [plan-14] Child Process Ownership — every process the worker spawns dies with the worker18 comments · 2026-08-16

Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.

05

Compatibility

nodejs
Node.js 20.19+ in 20.x, or 22.12+
Verified via docs
agent-hooks
Capture requires a supported agent adapter; some hosts are search-only
Verified via docs
06

Quick start

1
$ npx claude-mem install
07

Use cases

↳Recover project context in a later coding session.
↳Search past observations when a task needs details.
↳Share selected memory across supported agent integrations.
08

Alternatives

⚡Top VS:
VScognee★33.4k→VShermes-agent★264.2k→VSoh-my-openagent★71.0k→VSwidemem-ai★50→
cognee logo
cognee★ 33.4k
Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory with small models for free
VSCompare→
hermes-agent logo
hermes-agent★ 264.2k
The agent that grows with you
VSCompare→
oh-my-openagent logo
oh-my-openagent★ 71.0k
OmO: Just type "mass ulw" keyword with your prompt. Now you are the master of graph engineering.
VSCompare→
widemem-ai logo
widemem-ai★ 50
Next-gen AI memory layer with importance scoring, temporal decay, hierarchical memory, and YMYL prioritization
VSCompare→
TrueMemory logo
TrueMemory★ 385
A living memory system that ingests long-horizon data to infer insights, enabling more decisive action, all while running on a single SQLite file locally.
magic-context logo
magic-context★ 2.5k
Unbounded context. Memory that manages itself. One session, for life. The hippocampus for coding agents, part of CortexKit.
VSCompare→
swarmclaw logo
swarmclaw★ 712
Self-hosted multi-agent AI runtime with MCP server support. Build, orchestrate, and run autonomous agents with memory, skills, schedules, and 23+ LLM providers (Claude, GPT, Gemini, OpenRouter, Ollama).
origin logo
origin★ 498
Local-first Rust daemon for AI work. Git-versioned memories, distilled wiki pages, sessions for Claude Code, Cursor, Codex, and any MCP client.
VSCompare→
See all alternatives →

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Why choose it03Trade-offs04Maintenance check05Compatibility06Quick start07Use cases08Alternatives
thedotmack /
claude-mem
↗

Companion memory agent running alongside coding CLIs to index project notes and maintain long-term context across sessions

118.0kApache-2.0
Last commitToday
StatusActive
CategoryMemory & Context
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.