yizhiyanhua-ai

Fireworks AI Tools Memory — AI skill for Claude Code

AI community

Persistent cross-session memory for AI tools, CLI workflows, and reusable scripts across Codex and Claude-style sessions.

How to install Fireworks AI Tools Memory

This entry records only its repository, not the path inside it, so there is no exact command to give. Open yizhiyanhua-ai/fireworks-ai-tools-memory and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Fireworks AI Tools Memory does

Persistent cross-session memory for AI tools, CLI workflows, and reusable scripts across Codex and Claude-style sessions.

Alternatives in AI

Nocturne Memory — A lightweight, rollbackable, and visual Long-Term Memory Server for MCP Agents 1.3k ★
Pebrel — AI-native, GPU-accelerated terminal emulator for Windows with SSH, persistent sessions, split panes, and first 670 ★
Nebula — AI-native, GPU-accelerated terminal emulator for Windows with SSH, persistent sessions, split panes, and first 587 ★

README

fireworks-ai-tools-memory

**Persistent experience memory for AI tools, CLI workflows, and reusable scripts.**

This project extends the memory idea beyond skills and focuses on the messier operational layer: tools, toolchains, auth quirks, proxy rules, fallback paths, and scripts that already paid their tuition once.

[中文文档](README.zh-CN.md) · [MIT License](LICENSE)

![fireworks-ai-tools-memory landing image](assets/images/fireworks-ai-tools-memory-landing.png)

Natural-Language Installation

If you mainly work through Codex, the most natural install flow is to ask for it directly instead of starting from shell commands:

Install fireworks-ai-tools-memory
Add fireworks-ai-tools-memory to my current Codex environment
Make this a shared skill so all my Codex accounts can use it

The command-line path is just the fallback:

npx skills add yizhiyanhua-ai/fireworks-ai-tools-memory

If you already cloned the repository locally, you can also bootstrap it with:

./install-codex.sh

Common Natural-Language Requests

This skill is most useful when you want to stop relearning the same tool pain:

Inject the yt-dlp+mpv lessons before we start
Remember what went wrong with ncm-cli this time
Store this script in the tool memory library
Warn me about Spotify CLI auth pitfalls next time
Flush the tool lessons from this session

Typical CLI commands underneath:

python3 cli/tools_memory.py inject --tool yt-dlp+mpv
python3 cli/tools_memory.py checkpoint --tool ncm-cli --note "Search may work while playback still fails."
python3 cli/tools_memory.py flush --tool spotify-cli --summary-file ./session-summary.md
python3 cli/tools_memory.py register-script --tool yt-dlp+mpv --source ./scripts/play_mix.sh --name play-mix
python3 cli/tools_memory.py list-scripts --to