Agent Memory
Persistent cross-session memory using vector and graph search that automatically records and compresses every session's work, injecting only the relevant context into the next session.
Before & after
Same prompt, same agent — the only difference is the skill.
I'd need more context to help you find that decision. Can you give me some details about what kind of auth discussion you're referring to? Was it about OAuth, JWT, session management, single sign-on, or something else? What was the general context - were you talking about implementing a new auth system or debugging an existing one? If you can point me to the session or conversation where that happened, I could help you find the specific decision you're looking for.
AGENT MEMORY RETRIEVAL - AUTH DECISIONS VECTOR SEARCH RESULTS (relevance 0.94 - 0.87): 1. Session 47 (2026-06-28): OAuth 2.0 Migration Decision - DECISION: Migrate from custom auth to Auth0, 30-day deadline - Key factors: Support ticket volume down 60% with managed provider - Cost impact: +$400/mo, saves 12 eng-hours/week - Owner: Sarah (auth lead) - Status: In progress, 70% complete 2. Session 44 (2026-06-18): JWT Expiry Windows - DECISION: Set access tokens to 15 min, refresh to 7 days - Trade-off: Security vs. UX friction - Accepted risk: Refresh token theft (mitigated by device fingerprint) - Owner: Marcus - Status: Deployed to staging 3. Session 39 (2026-05-30): Passwordless Auth Trial - DECISION: Pilot magic links for creator onboarding only - Metrics: 40% faster signup, 8% bounce reduction - Rollout: Phase 2 starts July 15 - Owner: Priya - Status: Active, metrics tracked WEB VIEWER: View full decision graph and edit stored context at localhost:5002
About this skill
name: agent-memory description: Use when Persistent cross-session memory using vector and graph search that automatically records and compresses every session's work, injecting only the relevant context into the next session.
Agent Memory
Goes beyond basic markdown session-log memory, which becomes unmanageable as it grows and can't capture everything. Agent Memory silently records everything that happens across every session with no manual logging required — just code normally. It automatically compresses activity into a local database, then on the next session injects only the context relevant to the current task using a combination of vector search and graph search, instead of dumping an entire file into context (which pulls in irrelevant noise). A web viewer lets you directly inspect what Claude currently remembers about your project.
What you get
- Fully automatic session recording with zero manual logging, vector + graph search retrieval instead of full-file context dumping, web viewer to inspect what's remembered, scales far better than markdown-based memory as history grows.
Customize your output
- Adjust retrieval sensitivity for vector vs. graph search and review/edit what's stored via the web viewer.
Example output
On session 12, Agent Memory automatically pulled in only the 3 relevant past decisions about the auth refactor, instead of the entire multi-week session history.
Best for
Developers on long-running projects who've hit the limits of markdown-based session logs and want smarter, automatic context retrieval.
SKILL.md preview
---
name: agent-memory
description: Use this skill when a project needs persistent cross-session memory that automatically records and compresses work, instead of manually maintained markdown logs.
version: 1.0.0
category: Development / Context Management
author: AgentVolt
license: proprietary
tags:
- development
- context-management
- standard
---
# Agent Memory
Provides persistent cross-session memory using vector and graph search, silently recording every session's work and injecting only what's relevant into the next session.
## When to use
… (sign up to view the full skill)