Mem

@Mem create a note titled 'Q1 Product Roadmap' with bullet points for the three features we just discussed in this thread.

I've created a new note in Mem titled 'Q1 Product Roadmap' containing the three features from our conversation: advanced search filters, bulk export functionality, and mobile offline mode. The note is saved with a unique ID and timestamp. You can retrieve or edit it anytime using the note ID, or search for it by title in your Mem workspace.

This example shows Mem's core note-creation capability. The MCP supports idempotent creation via a memid parameter, so repeated calls with the same ID won't duplicate content. Requires a valid API key with write permissions. Notes are private to your Mem account unless you explicitly share collection access.

@Mem read the note with ID mem_abc123xyz and summarize the action items in priority order.

I retrieved the note 'Sprint Planning – Week 12' from Mem. Here are the action items in priority order: (1) Deploy hotfix for login timeout issue by EOD Thursday, (2) Review design mockups for settings page by Friday morning, (3) Schedule user testing sessions for next week, (4) Update API documentation with new endpoint examples. The note also contains meeting attendees and a link to the Figma file.

This example pairs Mem's read tool with AI reasoning to extract and reorder content. You must know the note ID in advance—Mem's MCP doesn't expose a search-by-keyword tool, so you'll need to track IDs externally or retrieve them from Mem's UI. The AI interprets the note's structure but doesn't modify the original.

@Mem create a collection called 'Customer Feedback' and add a note summarizing the three support tickets we reviewed earlier.

I've created a new collection in Mem titled 'Customer Feedback' and added a note summarizing the support tickets: recurring login errors on mobile (reported by 4 users), feature request for CSV export (3 users), and confusion about billing cycle dates (2 users). The collection is now available in your Mem workspace, and the note is linked to it. You can add more notes to this collection as new feedback arrives.

This example demonstrates Mem's organizational structure—collections group related notes. The MCP creates the collection first, then the note, requiring two API calls. If the collection name already exists, the create-collection call may fail depending on Mem's duplicate-handling logic. Always confirm collection titles before running this workflow to avoid errors.

When Mem wins for post-call knowledge capture

A 3-person consulting team runs 8-12 client calls a week and needs structured notes that surface later when context matters. Mem's collection-per-client model works here: each consultant creates notes during or after calls, tags them to the client collection, and the AI workspace can read those notes when drafting follow-up emails or proposals. The idempotent creation via memid means you can safely retry note saves without duplicates. The threshold: if your team needs real-time collaboration on the same note during a call, Mem's single-author model breaks down—you'd want a multiplayer doc tool instead. For solo capture that feeds team AI context, Mem is the right call.

When Mem fills gaps in lightweight deal tracking

A 6-person sales team uses a basic CRM but needs a scratchpad for unstructured deal intel—competitor mentions, stakeholder quirks, objection patterns—that doesn't fit CRM fields. Mem collections map to deals or accounts, and reps create notes as they learn. The AI workspace reads those notes when prepping for calls or writing proposals. The win is speed: creating a note in Mem takes one tool call versus navigating CRM custom fields. The trade-off: Mem has no search tool in this MCP, so if your team needs to query across 200+ deals for a pattern, you're reading notes one by one. Under 50 active deals, Mem's simplicity beats CRM bloat.

When Mem works for curated reference collections

A 4-person design team saves screenshots, color palettes, and layout ideas into Mem collections organized by project or theme. Each designer creates notes with image URLs and short annotations, and the AI workspace pulls examples when generating design briefs or mood boards. The collection structure keeps inspiration organized without folder hell. The constraint: Mem's 5 tools don't include search or list-all-notes, so discovery relies on knowing which collection to read. If your team's library grows past 20 collections or 300 notes, you'll hit friction finding the right reference. For teams under that scale who want fast capture and AI-readable context, Mem delivers.