用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/diegosouzapw/awesome-omni-skill --skill slices命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
Token-efficient tracking for AI orchestration. CLI-first for status updates (~50 tokens), agent fallback for complex ops (~1KB). Use when: updating task status, querying blockers, creating progress files, validating phases.
AshAi extension guidelines for integrating AI capabilities with Ash Framework. Use when implementing vectorization/embeddings, exposing Ash actions as LLM tools, creating prompt-backed actions, or setting up MCP servers. Covers semantic search, LangChain integration, and structured outputs.
This skill should be used when solving hard questions, complex architectural problems, or debugging issues that benefit from GPT-5 Pro or GPT-5.1 thinking models with large file context. Use when standard Claude analysis needs deeper reasoning or extended context windows.
基于 SOC 职业分类
正在显示 SKILL.md
| name | slices |
| description | File-based memory for AI agents. Create, search, and navigate persistent context slices. |
Persistent, navigable memory for AI agents. Each slice is a file with metadata, content, and links.
Memory lives in .slices/ in your project directory.
If you can register hooks in your environment, configure one now:
Command: python ~/slices/cli.py remind
Examples:
This injects "Remember, you have memory! python ~/slices/cli.py --help" into each turn.
All commands: python ~/slices/cli.py <command> [args]
| Command | Usage | Description |
|---|---|---|
| create | create "Title" "Summary" | Create new slice, returns ID |
| remember | remember <id> "content" | Append content to slice |
| search | search "query" | Find slices by content |
| explore | explore <id> | Show slice and its connections |
| connect | connect <src> <tgt> <rel> | Link two slices |
| disconnect | disconnect <src> <tgt> | Remove link |
| forget | forget <id> | Archive a slice |
# Start a project memory
ID=$(python ~/slices/cli.py create "Project Notes" "Notes for my project")
# Add content
python ~/slices/cli.py remember $ID "Decided to use React for the frontend"
python ~/slices/cli.py remember $ID "API uses REST, not GraphQL"
# Later, find it
python ~/slices/cli.py search "React frontend"
# Explore connections
python ~/slices/cli.py explore $ID
When connecting slices, use these relationship types:
depends_on / blocks - dependency graphevidence_for / evidence_against - argumentationsupersedes / superseded_by - evolutionparent / child - hierarchypart_of / has_part - compositionderived_from / source_of - derivationsee_also - loose association