agentleFS
Sign inStart free
naman10parikhnaman10parikh/wikimem/.claude/skills/memory/SKILL.md

memory

Skill8 starsChanged 26 days agoMITView on GitHub

Search, store, compress, and recall memories across the Energy platform. Semantic search over LEARNINGS.md, daily logs, vault notes, and rules.

99 other projects use a version of this, 3 of them edited.

Adopted by

Most recently taken up by

Used with

Projects with it also use send-user-message, docs-ci and 2 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • Maliot100XSkill

    send-user-message

    Send a proactive message to a paired user via their channel (Discord, Telegram, or WhatsApp). Use when the agent needs to notify, alert, or send an unsolicited message to a user — especially during heartbeat invocations, scheduled tasks, or when the agent wants to reach out without a prior user message in the current conversation. Triggers: 'send message to user', 'notify user', 'alert user', 'message the user on discord/telegram/whatsapp', or any need to proactively communicate with a paired sender.

    Also in 4 projects that use this

  • jh941213Skill

    docs-ci

    문서 CI(링크·OpenAPI lint·breaking 게이트·mermaid 검증)나 문서 드리프트 감지(docs.yaml 신선도 검사)를 설치하거나 돌릴 때 사용. 문서 내용 작성은 docs-architecture·docs-interfaces·docs-manuals.

    Also in 3 projects that use this

  • jh941213Skill

    docs-interfaces

    API·이벤트 인터페이스 문서(OpenAPI·swagger·AsyncAPI 명세, 엔드포인트 레퍼런스, API 구성도·시퀀스, API CHANGELOG)를 만들거나 코드와 맞출 때 사용. 시스템 구성도는 docs-architecture, API 설계 원칙은 api-design-principles.

    Also in 3 projects that use this

  • +1 more from jh941213

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 6 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 81 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Search across all memory surfaces with relevance scoring.

    line 15
  2. Distinctive:

    If no results, try broader search terms

    line 22
  3. Very specific:

    Use the memory MCP tool: create_entities with entity type based on content

    line 32
  4. Very specific:

    Run bash scripts/memory-compress.sh (if exists)

    line 44
  5. Distinctive:

    Keep recent 30 days in LEARNINGS.md

    line 47
  6. Distinctive:

    Add compressed index header with category counts

    line 48
  7. Very specific:

    Search memory surfaces (same as /memory search)

    line 57
  8. Distinctive:

    Check MCP memory server for stored entities matching topic

    line 59
  9. Distinctive:

    Before compaction: store critical facts via MCP for cross-session persistence

    line 93

“Common” is measured: kept by at least 6 in 10 of the 81 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/memory — Energy Memory System

Unified memory interface for the Energy platform. Search across all memory surfaces, store persistent facts, compress old entries, and recall context.

Commands

/memory search "query"

Search across all memory surfaces with relevance scoring.

Steps:

  1. Run bash scripts/memory-search.sh "query" --limit 5
  2. Parse results: file path + line number + surrounding context
  3. Present top 5 results ranked by: term frequency x recency x source weight
  4. If no results, try broader search terms

Sources searched: memory/LEARNINGS.md, memory/daily/.md, memory/topics/.md, vault/.md, .claude/rules/.md

/memory store "key" "value"

Store a persistent fact via MCP memory server.

Steps:

  1. Use the memory MCP tool: create_entities with entity type based on content
  2. For maintainer directives: entity type = "directive"
  3. For technical patterns: entity type = "pattern"
  4. For project state: entity type = "project"
  5. Confirm storage with entity name

/memory compress

Compress LEARNINGS.md by archiving old entries.

Steps:

  1. Run bash scripts/memory-compress.sh (if exists)
  2. Or manually: read LEARNINGS.md, identify entries older than 30 days
  3. Move old entries to memory/archive/learnings-YYYY-MM.md
  4. Keep recent 30 days in LEARNINGS.md
  5. Add compressed index header with category counts
  6. Report: lines before, lines after, entries archived

/memory recall "topic"

Recall everything known about a topic across all surfaces.

Steps:

  1. Search memory surfaces (same as /memory search)
  2. Also search: .claude/auto-memory files, vault notes, maintainer prompts
  3. Check MCP memory server for stored entities matching topic
  4. Combine and deduplicate results
  5. Present as structured brief: what we know, when we learned it, confidence level

Architecture (4 Layers)

Layer 0: MARKDOWN FILES (source of truth — always works)
  memory/LEARNINGS.md, memory/daily/, memory/topics/, vault/

Layer 1: SEARCH INDEX (scripts/memory-search.sh)
  BM25-style term frequency search with recency weighting
  Searches: LEARNINGS, daily, topics, vault, rules

Layer 2: MCP MEMORY SERVERS (persistent K-V + knowledge graph)
  @modelcontextprotocol/server-memory — entity-relation graph
  mixgram — document semantic search over memory/ directory

Layer 3: FLEET MEMORY (future — Mem0/Graphiti when agent fleet grows)
  Agent-to-agent knowledge propagation
  Temporal validity tracking

Key Principle

Layer 0 is ALWAYS the source of truth. Upper layers are indexes and caches that can be rebuilt from Layer 0 at any time. If any MCP server fails, grep still works.

When to Use This Skill

  • "What do we know about X?" → /memory recall "X"
  • "Find that learning about compaction" → /memory search "compaction"
  • "Remember this pattern for future sessions" → /memory store
  • "LEARNINGS.md is too long" → /memory compress
  • At session start: search for context relevant to current task
  • Before compaction: store critical facts via MCP for cross-session persistence

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free