- Home
- Skills
- Workflow Automation
- Agent Project Memory
Works with the AI Agents you already use
Agent Project Memory
Persistent, project-scoped memory for AI coding agents inside a hidden .memory/ repo folder.
$1
Agent Project Memory
Example session with this skill installed
Continue where we left off on the authentication middleware.
- Read your context and instructions
- Compiled the agent project memory
Read .memory/handoff.md. Found active task 'JWT token rotation' in src/auth.ts. Resuming implementation at line 42...
Connects securely to your tools. The creator never sees your data.
About this skill
About this skill
AI coding agents suffer from context resets—forgetting your tech stack, repeating old bugs, and losing track of multi-step tasks between sessions. Agent Project Memory gives any coding agent a persistent, git-safe .memory/ brain right inside your repository root.
What it does
- Session Wakeup: Reads
.memory/project.md,memory.md, andhandoff.mdat the start of every session to resume work with zero re-prompting. - Automatic Handoffs: Logs completed tasks, modified files, active blockers, and the exact next step before closing a session.
- Architecture Guardrails: Checks
memory.mdbefore adding libraries or changing schemas so agents never drift from your stack. - Git & Privacy Protection: Ships with a built-in
.memory/.gitignorethat keeps volatile session notes (handoff.md,scratch/) strictly untracked while keeping durable Architecture Decision Records (decisions/) versioned.
How it works
- Initialize: Creates a hidden
.memory/folder in your project root from clean Markdown templates. - Wakeup Check: On prompt start, the agent loads project context, stack rules, and the last session handoff.
- Guardrail Enforcement: Validates new code changes against recorded architectural principles in
memory.md. - Shutdown Sync: Rewrites
handoff.mdwith current state and appends durable lessons tomemory.mdandchangelog.md.
Why this beats prompting it yourself
Manually pasting context into every new chat wastes tokens and still misses edge cases. This skill enforces a deterministic file-based lifecycle across sessions and works seamlessly even when switching between different agents (Claude Code, Cursor, Antigravity, Gemini CLI).
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Workflow Automation.
Requires read/write access to the local project workspace. Does not sync across separate machines without git push.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 9 days ago
Nothing to install. It runs in your AI agent.
- Optional
Git
Permissions
File Scopes
Operates strictly on local markdown files inside .memory/. Zero network calls, zero external binaries, and built-in gitignore protection for session state.