# AI use cases for Claude Mem (thedotmack/claude-mem) Repo: [thedotmack/claude-mem](https://github.com/thedotmack/claude-mem) | Use cases: 4 | Page: https://dedupedai.com/use-cases/thedotmack-claude-mem/ Claude-Mem is a persistent memory system that captures what coding agents do in a session, compresses it with AI, and injects relevant context back into later sessions. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Carry Claude Code project context into new sessions: https://dedupedai.com/use-cases/thedotmack-claude-mem/resume-claude-code-project-context/ - Find past decisions and bug fixes before you repeat them: https://dedupedai.com/use-cases/thedotmack-claude-mem/search-past-decisions-in-claude-code/ - Give Codex threads in T3 Code shared project memory: https://dedupedai.com/use-cases/thedotmack-claude-mem/add-shared-memory-to-t3-code-codex-threads/ - Run persistent memory on an OpenClaw gateway with chat feeds: https://dedupedai.com/use-cases/thedotmack-claude-mem/feed-openclaw-gateway-memory-to-chat/ ## How to carry Claude Code project context into new sessions with Claude Mem Repo: [thedotmack/claude-mem](https://github.com/thedotmack/claude-mem) | Difficulty: Intermediate | Time: 15 minutes | Basis: From the README Install the Claude Code plugin so earlier session context appears automatically when you start a new session. **Who it is for:** A developer working on the same codebase with Claude Code across many days, who keeps re-explaining past decisions at the start of each session. **The problem:** Claude Code sessions end and the agent forgets what was built, decided, and fixed in earlier sessions. **The outcome:** New Claude Code sessions open with summaries of prior work on the project, so the agent continues where the last session stopped. ### Steps 1. In Claude Code, run `/plugin marketplace add thedotmack/claude-mem` 2. Run `/plugin install claude-mem` 3. Restart Claude Code 4. Start a new session in the same project and check that context from the previous session appears ### Prompt to paste into your agent ```text Install the claude-mem plugin in Claude Code for the repo at using `/plugin marketplace add thedotmack/claude-mem` and `/plugin install claude-mem`, then restart Claude Code. Report whether a new session in shows context from a previous session. Verify by checking that the injected context references work from the earlier session. ``` ### Caveat The README says the npm package installs only the SDK, so the plugin must be installed through the /plugin commands or npx Claude Mem install. Risk, credentials: Signing in through the installer provisions a memory key for your account, so review what the account and provider settings grant before you connect it. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/thedotmack-claude-mem/resume-claude-code-project-context/ ## How to find past decisions and bug fixes before you repeat them with Claude Mem Repo: [thedotmack/claude-mem](https://github.com/thedotmack/claude-mem) | Difficulty: Intermediate | Time: 15 minutes | Basis: From the README Use the mem-search tools to look up earlier bugfixes and decisions, then fetch only the relevant details. **Who it is for:** A developer or agent who must recall why something was built a certain way or how a bug was fixed weeks earlier in the same project. **The problem:** Searching raw session history is slow, and loading full transcripts costs a lot of tokens. **The outcome:** The agent finds the matching observations by ID, reviews a timeline around them, and loads full details only for the IDs it needs. ### Steps 1. Confirm the Claude Mem plugin is installed and that Claude Code was restarted 2. Ask Claude a natural language question about past work in the project 3. Call `search` with a query and an optional type filter, for example `search(query="authentication bug", type="bugfix", limit=10)` 4. Call `timeline` on an interesting result to see what happened around it 5. Call `get_observations` with the chosen IDs in one batch, for example `get_observations(ids=[123, 456])` ### Prompt to paste into your agent ```text Using the claude-mem MCP tools, search memory for in the project at , then use timeline on the most relevant result and get_observations with the chosen IDs in one batch. Report a short list of decisions or fixes with their IDs. Verify that every ID you report came from the search results. ``` ### Caveat The README describes token savings of about 10x from filtering before fetching, but the exact savings depend on your own results and are not guaranteed. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/thedotmack-claude-mem/search-past-decisions-in-claude-code/ ## How to give Codex threads in T3 Code shared project memory with Claude Mem Repo: [thedotmack/claude-mem](https://github.com/thedotmack/claude-mem) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Run the T3 Code installer so Codex and Claude Code providers in T3 Code get Claude Mem memory. **Who it is for:** A developer who runs Codex through T3 Code and wants new threads to start with context from earlier work. **The problem:** Each new Codex thread starts with no knowledge of the project's earlier sessions. **The outcome:** New T3 Code threads with Codex receive memory from prior sessions after the installer and hook trust steps are done. ### Steps 1. Run `npx claude-mem install --ide t3code` 2. Restart T3 Code 3. Trust the provider's hooks when T3 Code prompts you 4. Start a new thread and check that prior context appears ### Prompt to paste into your agent ```text Run `npx claude-mem install --ide t3code` on this machine, then restart T3 Code and start a new Codex thread in . Report which providers the installer discovered and whether the new thread shows prior context. Verify the installer output lists the Codex provider. ``` ### Caveat The README says custom server settings, status, and removal are covered in the T3 Code integration guide, which is outside this README. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/thedotmack-claude-mem/add-shared-memory-to-t3-code-codex-threads/ ## How to run persistent memory on an OpenClaw gateway with chat feeds with Claude Mem Repo: [thedotmack/claude-mem](https://github.com/thedotmack/claude-mem) | Difficulty: Advanced | Time: 30 minutes | Basis: From the README Install the Claude Mem plugin on an OpenClaw gateway and optionally send observation feeds to chat apps. **Who it is for:** An operator of an OpenClaw gateway who wants agent memory that persists and wants live observations sent to team chat. **The problem:** OpenClaw gateway agents lose context between sessions, and the team has no view of what the agent observed. **The outcome:** The gateway runs Claude Mem as a persistent memory plugin, and optional real-time observation feeds go to chat tools the installer supports. ### Steps 1. Confirm Node.js 20.0.0 or higher is installed on the gateway host 2. Run `curl -fsSL https://install.cmem.ai/openclaw.sh | bash` 3. Follow the installer prompts for AI provider configuration and optional feeds to Telegram, Discord, Slack, or others 4. Check that the worker started and that a test session produces an observation ### Prompt to paste into your agent ```text On the OpenClaw gateway at , run the claude-mem OpenClaw installer from https://install.cmem.ai/openclaw.sh, then confirm the worker is running. Report the provider chosen and any chat feed configured. Verify that a test session creates an observation in memory. ``` ### Caveat The README names Telegram, Discord, and Slack as feed targets and says more are supported, but it does not list the others, so check the OpenClaw integration guide before you rely on a specific one. Risk, remote script: The installer is a script piped from the network into bash, so download and read it first if your gateway policy requires review. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/thedotmack-claude-mem/feed-openclaw-gateway-memory-to-chat/