/memory
Update persistent swarm memory in ${SWARM_MEMORY_FILE} when the user explicitly asks to remember, update, or forget durable information.
$ npx -y skills add SawyerHood/middleman --skill memory --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
- You can call itInvoke it directly when you want it.
- Slash command
/memory
Context preview
The summary Claude sees to decide when to auto-load this skill.
Update persistent swarm memory in ${SWARM_MEMORY_FILE} when the user explicitly asks to remember, update, or forget durable information.
SKILL.md
memory.SKILL.mdname: memory
description: Update persistent swarm memory in ${SWARM_MEMORY_FILE} when the user explicitly asks to remember, update, or forget durable information.Persistent Memory Workflow
Use this skill when the user explicitly asks to:
- remember something for later,
- update previously remembered facts/preferences, or
- forget/remove stored memory entries.
Do not write memory for normal one-off requests.
File location
- Persistent memory files are stored at `${SWARM_DATA_DIR}/memory/<agentId>.md`.
- In this runtime, use `${SWARM_MEMORY_FILE}` (also shown in your loaded context).
Steps
1. Read the current memory file with `read` before changing it. 2. Apply minimal edits:
- prefer `edit` for targeted changes,
- use `write` only for full rewrites.
3. Keep entries concise, factual, and durable. 4. Never store secrets (passwords, API keys, tokens, private keys) or highly sensitive personal data. 5. If the request is ambiguous, ask a clarifying question before writing. 6. After updating memory:
- manager: confirm the update to the user via `speak_to_user`,
- worker: report the update back to the manager via `send_message_to_agent`.
A local-first multi-agent orchestration platform. One manager, many workers, zero tab-juggling. Middleman is pre-v1, experimental, and updated constantly. Expect breaking changes.
Repo: SawyerHood/middleman