docs: add Engram MCP memory server integration guide

[manaporkun]

Summary

• Adds documentation for connecting Engram as an MCP server for persistent, searchable memory
• Includes install instructions (Homebrew and Go) and a ready-to-use config.json example
• Tested on Pi Zero 2 W running PicoClaw + Engram as a personal AI advisor

Test plan

• Verify the new section renders correctly in the GitHub markdown preview
• Confirm the config.json snippet is valid JSON
• Test the Engram install commands on a fresh environment

🤖 Generated with Claude Code

[Copilot]

Pull request overview

Adds a new documentation section describing how to run the Engram MCP server alongside PicoClaw to provide persistent, searchable memory, including install commands and a config.json example.

Changes:

• Document Engram as an MCP server option for persistent memory.
• Add install instructions (Homebrew + go install).
• Provide a ready-to-use MCP server configuration snippet for config.json.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The example uses --project=my-assistant without indicating it’s a placeholder. To avoid users accidentally sharing/overwriting the same Engram project across environments, consider using a clearer placeholder (e.g., YOUR_PROJECT_NAME) and/or adding a short note that this value should be customized per assistant/user.

[Copilot]

This sentence hard-codes both the exact number of tools ("11") and specific tool names. Since these can change between Engram versions and also depend on the selected --tools= set, consider removing the exact count and phrasing it more defensively (e.g., “provides memory tools such as …”) while keeping the link to Engram’s tool reference.

Suggested change

	This gives your agent 11 memory tools including `mem_save`, `mem_search`, `mem_context`, and `mem_session_summary`. See the [Engram documentation](https://github.com/Gentleman-Programming/engram) for the full tool reference.
	This provides your agent with Engram memory tools such as `mem_save`, `mem_search`, `mem_context`, and `mem_session_summary`. See the [Engram documentation](https://github.com/Gentleman-Programming/engram) for the full tool reference.