memory
Decision intelligence system — agents learn decision rules from every task. Not a cache. Only stores what changes future behavior.
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。
菜单
Decision intelligence system — agents learn decision rules from every task. Not a cache. Only stores what changes future behavior.
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
Auto-checks project context, logs activity via scribe, and manages the activity log and artifact directories. Loaded by Optimus and Megatron on every command.
Silent logger skill. Provides Scribe agent with rules for activity logging, memory writes, and report generation within .claude/transformers/.
Deep git and VCS knowledge — worktrees, branching, rebase, conflict resolution, PR review, bisect, release workflow.
How to create effective Transformers skills — structure, progressive disclosure, bundled resources, testing.
Guided introduction to Transformers for new users — what it is, how to use it, what each agent does.
Pre-completion self-check for all builder agents. Run before marking any task done.
| name | memory |
| description | Decision intelligence system — agents learn decision rules from every task. Not a cache. Only stores what changes future behavior. |
| user-invocable | false |
You have access to project-specific decision memory that evolves over time. Memory changes how you act, not what you know.
Memory is NOT context. Facts about the project (stack, structure, files) belong in context/project.md. Memory stores decision rules — what to do, what to avoid, what works, what doesn't.
Check if .claude/transformers/memory/long-term/index.md exists
Check if .claude/transformers/memory/temp.md exists
If a rule in the index is relevant to your current task, read the full category file for the reasoning behind it.
When you had to figure something out the hard way — that's a memory candidate.
The test: Would knowing this change how an agent acts next time? If yes, save it. If no, skip it.
Tag it in your response to the orchestrator:
[MEMORY] brief decision rule — why
If you're a standalone command (no orchestrator), spawn Scribe directly:
"Append to
.claude/transformers/memory/temp.md:## [timestamp] [your-name] — [title]\n[decision rule and why]"
For rules that are clearly permanent, tag as:
[LONG-TERM] category: decision rule
project-context.mdstatus.mdactivity.logFull memory system rules: @${CLAUDE_PLUGIN_ROOT}/knowledge/memory-system.md