| name | 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
- Use
${SWARM_MEMORY_FILE} as the source of truth for the memory file path in this runtime (also shown in your loaded context).
- Do not derive memory paths manually from
${SWARM_DATA_DIR} or agent/session IDs; the runtime resolves the correct memory owner and path.
Steps
- Read the current memory file with
read before changing it.
- Apply minimal edits:
- prefer
edit for targeted changes,
- use
write only for full rewrites.
- Keep entries concise, factual, and durable.
- Never store secrets (passwords, API keys, tokens, private keys) or highly sensitive personal data.
- If the request is ambiguous, ask a clarifying question before writing.
- After updating memory:
- manager: confirm with the appropriate output path (normal final text for normal web/session chat, including a closeout reached from an internal worker result;
speak_to_user for collaboration, Cortex/review, non-web, or protected delivery),
- worker: include the update in the final assistant response, which Forge returns to the manager automatically.