- Home
- Skills
- Productivity
- Personal Memory for Codex — Your AI Context Companion
More screenshots
Works with the AI Agents you already use
Personal Memory for Codex — Your AI Context Companion
Maintain persistent, user-owned local memory profiles to keep your context and preferences across AI sessions.
$1.99
Personal Memory for Codex — Your AI Context Companion
Example session with this skill installed
I'm starting a new React project using Vite and Tailwind. Remember that I prefer using functional components with TypeScript and absolute imports for the @hooks directory.
- Read your context and instructions
- Compiled the personal memory
I have recorded these preferences to your local memory vault under the default profile. These details will be available in future sessions to ensure code generation matches your architectural style.
Memory Update: React Development Preferences
| Topic | Statement | Status | Source |
|---|---|---|---|
| tech-stack | Prefers React with Vite and Tailwind CSS. | active | User explicit |
| coding-style | Uses functional components with TypeScript. | active | User explicit |
| path-mapping | Uses absolute imports for the @hooks directory. | active | User explicit |
Next steps
- Run
capsulein your next project chat to inject these preferences. - Use
updateif you decide to switch to a different directory alias like@features. - Launch
serve --opento view or edit your full development profile in the browser.
Connects securely to your tools. The creator never sees your data.
About this skill
Your AI loses context the moment a chat ends. It forgets your tech stack, your coding preferences, and your project history, forcing you to repeat yourself every Monday morning.
This skill solves the "cold start" problem by maintaining a persistent, user-controlled memory vault. It provides a local SQLite-backed profile system that stores your background, preferences, and project context across sessions. Unlike platform-side memories, this is stored on your machine in plain text, giving you full control over what is remembered, updated, or forgotten.
What it does
- Persistent storage saves specific facts, preferences, and project context to a local SQLite database.
- Context capsules generate compact, budget-aware summaries of your profile to inject into new chat sessions.
- Profile management organizes information into distinct profiles like "work", "personal", or specific project roles.
- Local management UI launches a loopback-only web interface to search, edit, or delete records manually.
- Privacy controls handle sensitive record flags, expiry dates for temporary plans, and explicit deletion of revision history.
How it works
- Locate and init resolves the local storage directory and initializes the SQLite vault on your machine.
- Recall and capture uses commands to search active records or save new information from the current conversation.
- Review and prune manages the lifecycle of facts, moving imports to pending status and purging expired records.
- Export and port generates portable capsules or full backups to move your context between different AI providers.
Frameworks & tools
The skill utilizes a standard-library Python helper script and SQLite for local data persistence. It adheres to XDG base directory specifications for cross-platform compatibility on Linux, macOS, and Windows.
Why this beats prompting it yourself
Manual system prompts grow bloated and eventually hit context limits. This skill dynamically retrieves only the relevant facts for your current task, keeping the context window clean while ensuring the AI actually remembers your specific architectural decisions and workflow preferences.
Use cases
- Environment sync by storing specific CLI flags, paths, and tool preferences you use daily.
- Project onboarding to quickly brief an AI on existing codebase conventions and architectural rules.
- Learning tracking to maintain a history of concepts you have mastered and topics you are currently studying.
- Workflow persistence by saving complex multi-step instructions that you don't want to re-type every session.
Known limitations
Does not support automatic background collection of all chat data. Storage is plain text without native encryption. Requires local Python execution permissions to access the SQLite vault.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Productivity.
Requires local Python environment. No native encryption for the SQLite vault. Does not automatically capture all chat history; requires explicit triggers or summaries.
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 20 days ago
- Software
Python
Permissions
File Scopes
Reads and writes a user-selected local memory archive. The optional management interface runs on localhost. Importing or exporting records requires access to the files selected by the user. The archive is stored as plaintext SQLite. When a cloud AI model is used, information read into the conversation may be sent to the model provider.