本文へ移動
cccskills
無料GitHub で公開

consolidate-memory

Reflective pass over your memory files — merge duplicates, fix stale facts, prune the index.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md1.9 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Memory Consolidation

You're doing a reflective pass over what you've learned about this user and their work. The goal: a future session should be able to orient quickly — who they work with, what they're focused on, how they like things done — without re-asking.

Your system prompt's auto-memory section defines the directory, file format, and memory types. Follow it.

Phase 1 — Take stock

  • List the memory directory and read the index (MEMORY.md)
  • Skim each topic file. Note which ones overlap, which look stale, which are thin.

Phase 2 — Consolidate

Separate the durable from the dated. Preferences, working style, key relationships, and recurring workflows are durable — keep and sharpen them. Specific projects, deadlines, and one-off tasks are dated — if the date has passed or the work is done, retire the file or fold the lasting takeaway (e.g. "user prefers X format for launch docs") into a durable one.

Merge overlaps. If two files describe the same person, project, or preference, combine into one and keep the richer file's path.

Fix time references. Convert "next week", "this quarter", "by Friday" to absolute dates so they stay readable later.

Drop what's easy to re-find. If a memory just restates something you could pull from the user's calendar, docs, or connected tools on demand, cut it. Keep what's hard to re-derive: stated preferences, context behind a decision, who to go to for what.

Phase 3 — Tidy the index

Update MEMORY.md so it stays under 200 lines and ~25KB. One line per entry, under ~150 chars: - [Title](file.md) — one-line hook.

  • Remove pointers to retired memories
  • Shorten any line carrying detail that belongs in the topic file
  • Add anything newly important

Finish with a short summary: how many files you touched and what changed.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

ai-sdk

無料

Vercel AI SDK expert guidance. Use when building AI-powered features — chat interfaces, text generation, structured output, tool calling, agents, MCP integration, streaming, embeddings, reranking, image generation, or working with any LLM provider.

日本語の概要は準備中です。原文の説明を表示しています。

garochee33/DSH42026年10月9日 更新

Algorithm design and analysis skill for DSH; use when working on optimization, data structures, or computational procedures.

日本語の概要は準備中です。原文の説明を表示しています。

garochee33/DSH42026年10月9日 更新

Design and implement scalable API gateways, RESTful APIs, GraphQL endpoints, WebSocket handlers, and microservice communication patterns. Use when creating API routes, designing endpoints, implementing middleware, or setting up service mesh architectures.

日本語の概要は準備中です。原文の説明を表示しています。

garochee33/DSH42026年10月9日 更新

best-of-n

無料

Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner. Use when asked to "best of n", "try multiple approaches", "parallel implementations", "/best-of-n", or "/bon".

日本語の概要は準備中です。原文の説明を表示しています。

garochee33/DSH42026年10月9日 更新

check

無料

Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness. Use when asked to "check work", "verify changes", "self-verify", "/check", "/verify", "/check-work", or "/self-verify".

日本語の概要は準備中です。原文の説明を表示しています。

garochee33/DSH42026年10月9日 更新

CI/CD pipeline design, GitHub Actions workflows, deployment automation, infrastructure as code, and release management. Use when setting up CI/CD, automating deployments, configuring build pipelines, or managing releases.

日本語の概要は準備中です。原文の説明を表示しています。

garochee33/DSH42026年10月9日 更新

garochee33 のスキルをすべて見る

このスキルの問題を報告する