| name | save-to-mags |
| description | Save session knowledge to Mags' brain (mags Supermemory container). Use for session summaries, corrections, decisions, project state — anything that helps Mags persist between sessions. NOT for published canon (use /save-to-bay-tribune for canon). |
| version | 1.1 |
| updated | "2026-04-28T00:00:00.000Z" |
| tags | ["infrastructure","active"] |
| allowed-tools | Bash(node *), Bash(source *) |
| effort | low |
| disable-model-invocation | true |
Save to Mags
Save session knowledge to the mags Supermemory container.
This is Mags' brain. Session summaries, corrections, decisions, editorial thinking, project state. Content that helps the next session start smarter.
Do NOT use for: Published editions, rosters, canon content. Those go to bay-tribune via /save-to-bay-tribune or node scripts/ingestEdition.js. (/supermemory-save writes to super-memory — the junk drawer — not bay-tribune. Don't confuse them.) See [[../../../docs/SUPERMEMORY|SUPERMEMORY]] §Search/save matrix for the full container map.
Step 1: Format Content
Tag corrections separately from decisions — corrections carry more weight across sessions.
[SAVE:mike:<date>]
<What happened / what was decided>
<Key corrections if any — prefix with [CORRECTION]>
<Files changed if relevant>
[/SAVE]
Step 2: Save
source ~/.bashrc && curl -s -X POST "https://api.supermemory.ai/v3/documents" \
-H "Authorization: Bearer $SUPERMEMORY_CC_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"content": "FORMATTED_CONTENT",
"containerTags": ["mags"],
"metadata": {"type": "session_save", "date": "YYYY-MM-DD"}
}'
Step 3: Confirm
Print the returned document ID and confirm it saved to mags.