easy-agent-mem 0.4.2

CLI-first persistent memory for AI coding sessions with direct Obsidian note writing and local fallback storage.

agent-mem

CLI-first persistent memory for AI coding sessions.

agent-mem keeps coding context outside the live chat window. It saves structured summaries directly into an Obsidian vault when one is configured, or into a local fallback store inside the project when Obsidian is not available.

That gives you a simple workflow:

• save progress as memory
• reopen memory later with a focused query
• keep chat context smaller
• stop repeating the same project decisions across sessions

Why This Exists

Long coding chats drift. Important decisions get buried, context windows bloat, and new sessions start cold.

agent-mem fixes that with a local, inspectable memory layer:

• Obsidian-first when you want graph view, backlinks, and durable notes
• project-local fallback when you want zero extra setup
• CLI-first so the product works even without MCP or IDE-specific integrations

Core Features

• direct Obsidian vault connection during init
• Obsidian-friendly session notes with:
  • YAML frontmatter
  • wiki-links
  • Index.md updates
• local fallback in .agent-memory/memory.md
• CLI commands to:
  • initialize storage
  • save structured summaries
  • recall relevant memory
  • watch file activity and create automatic checkpoints
• optional MCP support for IDEs that expose tools reliably
• generated instruction files tailored for:
  • Cursor
  • Claude / VS Code
  • Antigravity
  • OpenCode

Installation

pip install easy-agent-mem

Optional MCP support:

pip install "easy-agent-mem[mcp]"

Quick Start

1. Initialize the project

agent-mem init

During setup you can:

• connect an Obsidian vault
• or skip it and use local fallback storage

2. Save a session summary

agent-mem summarize --summary "## Goal

Ship the release.

## Outcome

Prepared the package for publication.

## Key decisions
- Use Obsidian as primary storage.

## Files changed
- src/agent_mem/memory.py

## Open tasks or blockers
- Publish 0.4.2

## Next prioritized steps
- Verify the built wheel."

3. Recall memory later

agent-mem recall "release status"

4. Run automatic checkpointing

agent-mem watch

This monitors file activity in the current repo and writes an automated checkpoint once activity settles.

Storage Modes

Obsidian Mode

If you provide a vault path during agent-mem init, notes are written directly into:

<vault>/Memory/Agent-Mem/

Current layout:

<vault>/
  Memory/
    Agent-Mem/
      Index.md
      project-name-YYYY-MM-DD_HH-MM-session.md

Session notes are normal Markdown files, so Obsidian picks them up automatically.

They include:

• YAML frontmatter
• wiki-links like [[project-name]]
• file links like [[File - src/agent_mem/cli.py]]
• task links like [[Task - ship release]]

Local Fallback Mode

If you skip Obsidian, memory is stored in:

.agent-memory/memory.md

This keeps the product usable even in plain local repos without extra tools.

Commands

agent-mem init
agent-mem summarize --summary "..."
agent-mem summarize --summary-file session.md
cat summary.md | agent-mem summarize --stdin
agent-mem recall "auth decisions"
agent-mem watch
agent-mem status
agent-mem setup
agent-mem setup-vscode
agent-mem print-mcp-json
agent-mem serve

IDE Integration

agent-mem works best when the IDE is instructed to treat saved memory as the source of truth.

The repo can generate instruction files automatically for:

• AGENT-MEM-RULES.md
• .cursor/rules/agent-mem.mdc
• .claude/instructions.md
• CLAUDE.md
• .antigravity/rules.md
• .opencode/instructions.md

Recommended IDE workflow:

1. run agent-mem init
2. add the generated instructions to your IDE
3. use agent-mem summarize after milestones
4. use agent-mem recall at the start of new sessions

Optional MCP workflow:

agent-mem setup-vscode
agent-mem serve

Automated Checkpoints

agent-mem watch is a lightweight file-activity watcher.

It:

• polls the repo for file changes
• waits for a quiet period
• writes an automatic checkpoint summary after enough changes accumulate

Useful options:

agent-mem watch --interval 2 --quiet-seconds 20 --min-changes 3
agent-mem watch --once

This is not a replacement for a high-quality manual summary. It is a safety net to avoid losing context between work bursts.

Release Notes

0.4.1

• added tailored IDE-specific instruction files for Cursor, Claude, Antigravity, and OpenCode
• strengthened generated rules with explicit summarization checks and anti-hallucination guidance
• aligned the instruction-file generator with the checked-in templates

0.4.0

• introduced Obsidian-first storage with direct vault note writing
• added YAML frontmatter, wiki-links, and Index.md
• added CLI-first summarize and recall
• kept local fallback mode intact

Project Links

• PyPI: easy-agent-mem
• Source: github.com/atharvavdeo/agent-mem

License

MIT