菜单
Store a memory in erinra for future reference
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
Run parallel code review agents and consolidate findings into a unified report
Capture a follow-up idea or task as a work-item draft without interrupting current work. Use to quickly jot down something you think of mid-task so it gets tracked. Works with nibs, GitHub, Azure DevOps, or a Markdown fallback.
Take an under-specified work item and make it actionable — resolve its open questions through a short interview grounded in the code, and give it acceptance criteria. Use on a captured draft, or any item too vague to start on.
Reconcile what was built against what was planned, record decisions and deviations, close the item (a single phase or a whole plan), and file follow-ups for deferred work. Use after finishing a phase or plan to keep planned and actual from drifting apart.
Orchestrate execution of MULTIPLE nibs in one run. Selects a queue, understands the nibs collectively (including how they fit together), then chooses the best execution mechanism per cluster — single series agent, parallel fan-out, scripted workflow, or agent team — and dispatches with ONE approval gate. Use when the user wants to work several nibs together (in parallel or series) rather than one at a time. Complements /decaf-build:auto-dev and /decaf-build:auto-tdd (which handle a single nib).
Direct development with automated review. Plans implementation, executes via subagent, then auto-reviews. Use for work that isn't test-driven (UI, config, styling, infrastructure, scaffolding).
| name | remember |
| description | Store a memory in erinra for future reference |
| argument-hint | [what to remember] |
Store a memory in erinra.
Check that mcp__erinra__store is available in your tools. If not, tell the user:
Erinra MCP server is not connected. Run
claude mcp add erinra -- erinra serve -s userand restart Claude Code.
/decaf-memory:remember [description of what to remember]
Clarify if needed — If the user's input is vague, ask what they want to remember and why it matters.
Classify the memory — Choose a type:
preference — User preference or working stylepattern — A reusable approach or conventiondecision — An architectural or design choicebug-fix — A bug and its solutionfact — A piece of factual knowledgeconcept — A concept or mental modelnote — A general noteevent — Something that happenedStore it using mcp__erinra__store:
mcp__erinra__store({
content: "[the memory content]",
type: "[type]",
projects: ["[project-name if applicable]"],
tags: ["[relevant tags]"],
links: [{ target_id: "[id]", relation: "[relation]" }] // optional: link to related memories
})
The links parameter is optional. Use it to connect the new memory to existing ones with relations like related_to, caused_by, context_for, or supersedes.
Check the response — store returns the top-3 most similar existing memories. If a very similar one exists:
mcp__erinra__update on the existing memory insteadmcp__erinra__merge to combine themConfirm — Tell the user what was stored and how it was classified.
User: /remember HttpClient timeout doesn't work as expected with Polly retry
Response:
I'll store this. Can you clarify:
1. What was the unexpected behavior?
2. What was the solution?
[After user responds]
Stored as bug-fix:
- HttpClient.Timeout is per-attempt, not total
- Polly retry can exceed expected total timeout
- Solution: Wrap retry policy with Polly timeout policy
Tags: polly, httpclient, dotnet
User: /remember I prefer tabs over spaces
Response:
Stored as preference: "User prefers tabs over spaces for indentation"
Guide users away from storing: