菜单
Fuse insights from multiple knowledge sources (code, docs, memory, research, web) into coherent understanding. Use when you have scattered information that needs integration.
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。
基于 SOC 职业分类
Guide for safely discovering and installing skills from external repositories. Use when a user asks for something where a specialized skill likely exists (browser testing, PDF processing, document generation, etc.) and you want to bootstrap your understanding rather than starting from scratch.
Connect to MCP (Model Context Protocol) servers and create skills for repeated use. Load when a user wants to use an MCP server, connect to external tools via MCP, or when they mention MCP, model context protocol, or specific MCP servers.
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Letta Code's capabilities with specialized knowledge, workflows, or tool integrations.
Decomposes and reorganizes agent memory blocks into focused, single-purpose components. Use when memory has large multi-topic blocks, redundancy, or poor organization.
Comprehensive guide for initializing or reorganizing agent memory. Load this skill when running /init, when the user asks you to set up your memory, or when you need guidance on creating effective memory files.
Send messages to other agents on your server. Use when you need to communicate with, query, or delegate tasks to another agent.
| name | synthesize |
| description | Fuse insights from multiple knowledge sources (code, docs, memory, research, web) into coherent understanding. Use when you have scattered information that needs integration. |
| argument-hint | ["topic to synthesize"] |
| user-invocable | true |
| allowed-tools | Read, Glob, Grep, Bash, WebFetch, WebSearch |
Synthesize knowledge on this topic: $ARGUMENTS
Identify relevant knowledge sources across these categories:
| Source Type | Where to Look |
|---|---|
| Code | Current codebase (Glob/Grep), implementation patterns, tests, comments |
| Documentation | Specs, READMEs, ADRs, API docs in the repo |
| Memory | .claude/ project memory, agent memory, previous learnings |
| External research | WebSearch, Exa for papers, blog posts, community discussion |
| Issue history | bd list, closed issues, decision rationale |
Scan at least 3 source types. Don't rely on a single category.
For each source, extract:
Three integration strategies, used as appropriate:
Convergent — Multiple sources agree. High confidence. Extract the common principle.
Divergent — Sources disagree. Analyze why (different contexts? different values? different data?). Don't force resolution — document the conditions under which each view applies.
Complementary — Sources cover different facets. Weave into a complete picture, noting which source contributed which piece.
## Synthesis: [Topic]
### Sources Consulted
- [Source type]: [what was found]
### Convergent Findings (high confidence)
- [Finding]: supported by [sources]
### Divergent Findings (context-dependent)
- [Claim A] (applies when [X]) vs. [Claim B] (applies when [Y])
### Gaps
- [What no source addressed]
### Integrated Understanding
[Coherent narrative combining all sources — the actual synthesis]
### Confidence Assessment
[Overall confidence and what would increase it]
### Actionable Next Steps
[What to do with this understanding]