AgentIndex icon
AgentIndex
AgentRecall-MCP logo
AgentRecall-MCP
★ 371
vs
locus logo
locus
★ 15

AgentRecall-MCP vs locus

AgentRecall-MCP: AgentRecall is a learning loop for AI agents that provides persistent, compounding memory. It captures corrections automatically, surfaces past insights across projects, and uses a five-layer memory pyramid with Ebbinghaus decay and Bayesian feedback. Zero cloud, all local markdown files.; locus: Locus provides persistent project-aware memory for AI coding tools, using three memory layers: structural, semantic, and episodic. It works with any MCP-compatible client and offers 14 tools and 3 resources for exploring, searching, and managing memory. Carbon Copy enables passive conversation capture for later recall.

01

TL;DR

AgentRecall-MCP logo
AgentRecall-MCPMemory & Context · ★ 371

Maintain context across AI agent sessions (Claude Code, Cursor, etc.)

Read AgentRecall-MCP full review →
locus logo
locusMemory & Context · ★ 15

Maintain project context across AI coding sessions to avoid losing architectural decisions and file structure awareness

Read locus full review →
02

Side-by-Side Comparison

Field
AgentRecall-MCP logoAgentRecall-MCP
locus logolocus
Category
Memory & Context
Memory & Context
Stars
★ 371
★ 15
License
MIT
MIT
Install Command
claude mcp add --scope user agent-recall -- npx -y agent-recall-mcp
npx -y locus-memory@latest install codex
Updated
1w ago
4mo ago
Open Source
Yes
Yes
Website
↗ Visit
↗ Visit
GitHub
↗ GitHub
↗ GitHub
Tags
agent-memory, ai-agents, claude-code
ai-memory, claude-code, codex-cli
03

Features

AgentRecall-MCP logoAgentRecall-MCP
01Persistent, compounding memory with 200-line awareness cap
02Automatic correction capture and alignment checking
03Cross-project insight recall via keyword and semantic (pgvector) search
04Zero cloud, all local markdown files, Obsidian-compatible
0510 MCP tools for agents, plus SDK and CLI
locus logolocus
01Three memory layers: structural (auto-parsed), semantic (user-curated), episodic (auto-captured)
02Carbon Copy passive conversation capture via inbox-based event protocol
0314 MCP tools for exploring, recalling, searching, remembering, and managing memory
04FTS5 full-text search across all memory layers and conversation events
054-layer security: metadata-only, file denylist, content redaction, audit UX
04

Use Cases

AgentRecall-MCP logoAgentRecall-MCP
↳Maintain context across AI agent sessions (Claude Code, Cursor, etc.)
↳Capture and learn from user corrections in software development
↳Coordinate memory across multiple parallel agents
locus logolocus
↳Maintain project context across AI coding sessions to avoid losing architectural decisions and file structure awareness
↳Store and retrieve important decisions and their rationale for consistent development practices
↳Passively capture conversation history from coding sessions for later search and recall
FAQ

FAQ

What is the difference between AgentRecall-MCP and locus?
Both AgentRecall-MCP and locus are in the Memory & Context category. AgentRecall-MCP has 371 stars, while locus has 15 stars.
Which is better, AgentRecall-MCP or locus?
The best choice depends on your use case. Choose AgentRecall-MCP if Maintain context across AI agent sessions (Claude Code, Cursor, etc.), and locus if Maintain project context across AI coding sessions to avoid losing architectural decisions and file structure awareness.
Is AgentRecall-MCP free or open source?
Yes, AgentRecall-MCP is open source on GitHub (MIT).
Is locus free or open source?
Yes, locus is open source on GitHub (MIT).
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
→

Related

Alternatives to AgentRecall-MCP →Alternatives to locus →AgentRecall-MCP details →locus details →