SkillAgentSearch skills...

gemini-mem

Persistent memory extension for Gemini CLI with local SQLite storage, hook-based recall/write, and optional MCP tools

Install / Use

claude mcp add djinn-soul -- npx -y github:djinn-soul/gemini-mem

If the server publishes to npm under a different name, use that package instead — check the repo README.

About this skill
🔌

MCP Server

Model Context Protocol server

Quality Score

71/100

Supported Platforms

Claude Code
Claude Desktop
Gemini CLI

Tags

Our assessment of gemini-mem

gemini-mem scores 71/100 on our quality scale, 367th of 420 Data & Analytics skills we index.

Its MCP Server is 3.4 KB long, well organised into 11 sections with 7 code examples: a solid amount of guidance for an agent.

It has 3 GitHub stars, so there is little community track record yet; judge it on its content.

Substance
26/30
Structure
20/20
Description
12/15
Adoption
3/20
Freshness
11/15

Maintenance, license and trust

  • The repository was last updated about 6 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.
  • Our last check on 2026-09-26 found the source still online.
  • It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 90/100, with 1 caution from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-09-26. Automated pattern scan on 2026-09-26. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

gemini-mem compared with similar skills

All 4 of these similar skills score higher than gemini-mem; compare them before choosing.

SkillScoreStarsUpdatedFormat
gemini-mem (this skill)by djinn-soul7136mo agoMCP Server
claude-memby thedotmack10095.0ktodayCLAUDE.md
Agent-Reachby Panniantong10086.3k14d agoCLAUDE.md
headroomby headroomlabs-ai10074.1ktodayCLAUDE.md
rufloby ruvnet10073.6ktodayCLAUDE.md

Frequently asked questions

How do I install gemini-mem?
Run claude mcp add djinn-soul -- npx -y github:djinn-soul/gemini-mem. The install tabs above show the steps for each supported agent.
Which AI agents does gemini-mem work with?
It is written for Claude Code, Claude Desktop and Gemini CLI, as a MCP Server file. Other agents that read the same format can often use it too.
Is gemini-mem safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is MIT-licensed and scores 90/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is gemini-mem still maintained?
The repository was last updated about 6 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.

Gemini Memory Extension

Gemini CLI extension that saves and retrieves project memory from local SQLite, with MCP tools that share the same memory DB.

Official Gemini docs this README follows:

  • Writing extensions: https://geminicli.com/docs/extensions/writing-extensions/
  • Releasing extensions: https://geminicli.com/docs/extensions/releasing/

What You Get

  • Hook-based memory capture (AfterAgent) and retrieval (BeforeAgent, SessionStart)
  • Slash commands for memory inspection and cleanup
  • Optional MCP server tools for external agents/clients
  • Local per-project storage by default:
${HOME}/.gemini/gemini-mem/<repo-folder>/memory.db

Quick Start (Use This Extension)

npm install
npm run build
gemini extensions link .
gemini extensions list

Then restart Gemini CLI.

Daily Dev Loop

  1. Edit code.
  2. Rebuild: npm run build
  3. Test: npm test
  4. Restart Gemini CLI to pick up changes.

Project Layout

src/
  hooks/      # SessionStart, BeforeAgent, AfterAgent
  memory/     # sqlite store, schema, retrieval, dedupe, redaction
  gemini/     # summarizer/reranker prompts + validators
  commands/   # mem-status, mem-search, mem-last, mem-prune
  mcp/        # stdio and HTTP MCP servers + tool handlers
  runtime/    # process and Gemini CLI adapters
  cli/        # db init/migration scripts

Slash Commands

  • /mem:status
  • /mem:search <query>
  • /mem:last [limit]
  • /mem:prune [maxAgeDays] [importanceFloor]

Local equivalents:

npm run mem:status
npm run mem:status -- --json
npm run mem:search -- "query text"
npm run mem:last -- 5
npm run mem:last -- 5 --json
npm run mem:prune -- 30 2

MCP Tools

MCP is decoupled from hook runtime in this extension build. Hook execution does not require MCP server discovery.

  • memory_status
  • memory_get_context
  • memory_search
  • memory_save_observation
  • memory_cite
  • memory_end_session

Run transports:

npm run mcp:server
npm run mcp:http

If you want Gemini CLI to auto-discover this MCP server, register it explicitly in your user/project Gemini MCP settings.

Default HTTP endpoint:

http://127.0.0.1:3303/mcp

Recommended MCP usage order:

  1. memory_get_context
  2. memory_save_observation
  3. memory_end_session

Config

Settings are defined in gemini-extension.json (DB path, model, limits, timeouts, project mode, command override).

Opt-in hook telemetry:

  • Set MEM_HOOK_TELEMETRY=true
  • Writes JSONL events and failures to <project-db-folder>/hook.log (for example ${HOME}/.gemini/gemini-mem/<repo>/hook.log)

Release Checklist (What To Do)

This repo already has workflows in .github/workflows/ and release notes in PUBLISHING.md.

  1. Update version in both files:
    • package.json
    • gemini-extension.json
  2. Validate locally:
    • npm run build
    • npm test
  3. Commit and tag:
    • git tag vX.Y.Z
    • git push origin main --tags
  4. Let GitHub Actions create/publish the release artifacts.
  5. Install test from users' perspective:
    • gemini extensions install <repo-url>
    • or gemini extensions install <repo-url> --ref=vX.Y.Z

For gallery discoverability (from official docs):

  • Keep repository public.
  • Add GitHub topic gemini-cli-extension.
  • Keep gemini-extension.json at repository root.

Verification Commands

npm run build
npm test
npm run test:hooks
npm run test:mcp
npm run test:mcp-http

Contributing

See CONTRIBUTING.md.

Related Skills

View on GitHub
GitHub Stars3
CategoryData
Updated5mo ago
Forks0

Languages

TypeScript

Trust signals

90/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

1 low1 info