claude-mem

v13.5.0 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 1mo AI Coding Tools

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

ai ai-agents ai-memory anthropic artificial-intelligence chromadb

+14 more

claude claude-agent-sdk claude-agents claude-code claude-code-plugin claude-skills embeddings long-term-memory mem0 memory-engine openmemory llm sqlite supermemory

Affected surfaces

auth

Summary

AI summary

Updates Also in this release, PostHog, and https://claude.com/claude-code across a mixed release.

Changes in this release

Type	Severity	Summary	CVE
Feature
Feature	Low	Adds anonymous usage analytics telemetry by default. Adds anonymous usage analytics telemetry by default. Source: llm_adapter@2026-06-10 Confidence: high	—
Feature	Low	Adds live progress display during dependency installation steps. Adds live progress display during dependency installation steps. Source: llm_adapter@2026-06-10 Confidence: high	—
Feature	Low	Adds post‑install consent prompt for telemetry opt‑in/opt‑out. Adds post‑install consent prompt for telemetry opt‑in/opt‑out. Source: llm_adapter@2026-06-10 Confidence: high	—
Feature	Low	Adds `npx claude-mem telemetry [status\|enable\|disable]` CLI commands for managing telemetry. Adds `npx claude-mem telemetry [status\|enable\|disable]` CLI commands for managing telemetry. Source: llm_adapter@2026-06-10 Confidence: low	—
Feature	Low	Ensures worker shutdown flushes telemetry within a hard 3‑second bound. Ensures worker shutdown flushes telemetry within a hard 3‑second bound. Source: llm_adapter@2026-06-10 Confidence: low	—
Feature	Low	Provides CLI commands to manage telemetry status (enable/disable/status). Provides CLI commands to manage telemetry status (enable/disable/status). Source: granite4.1:30b@2026-06-10-audit Confidence: low	—

Full changelog

Anonymous usage analytics (PostHog) — and the v13.5.0 release

claude-mem now ships anonymous, privacy-hardened usage analytics. This is the first release with any telemetry, and it follows the standard dev-tool model (Homebrew, Next.js, Astro): on by default, one command to opt out, and incapable of carrying your content by construction.

What's collected

Eight events (install_completed, install_failed, uninstall_completed, worker_started, session_compressed, context_injected, search_performed, error_occurred), identified by a random install UUID generated locally. Every property passes a strict whitelist scrubber — only numbers, booleans, and values from closed sets we define (platform, version, IDE choice, durations, counts) can leave your machine.

Never collected — enforced by whitelist, not blocklist: prompts or conversation content, file paths, source code, project or repo names, search queries, error messages, IP addresses, hardware identifiers, env values, emails, or any PII.

Opting out

Any one of these turns it off:

npx claude-mem telemetry disable
DO_NOT_TRACK=1 (the universal standard — overrides everything)
CLAUDE_MEM_TELEMETRY=0

npx claude-mem telemetry status shows the current state and which setting decided it. The installer asks once at the end of npx claude-mem install, and your answer is never re-asked.

Full documentation of every field and event: https://docs.claude-mem.ai/telemetry

Also in this release

Install flow: live progress for dependency steps and a consent prompt at the end of install
npx claude-mem telemetry [status|enable|disable] CLI command
Worker shutdown now flushes telemetry with a hard 3s bound — never delays stop

🤖 Generated with Claude Code

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track claude-mem

Get notified when new releases ship.

About claude-mem

A Claude Code plugin that automatically captures everything Claude does during your coding sessions, compresses it with AI (using Claude's agent-sdk), and injects relevant context back into future sessions.