mem-search

Memory Search

Search past work across all coding sessions using the agent-memory MCP tools.

When to Use

Use when users ask about previous sessions (not the current conversation):

• "Did we already fix this?"
• "How did we solve X last time?"
• "What happened last week?"
• "What decisions did we make about the auth system?"

3-Layer Workflow (ALWAYS Follow)

NEVER fetch full details without filtering first. 10x token savings.

Step 1: Search — Get Index with IDs

Use the search MCP tool (server: agent-memory):

search(query="$ARGUMENTS", limit=20)

Returns: table with IDs, titles, types, scores (~50-100 tokens/result)

Optional filters:

Param	Example	Description
project	"myapp"	Filter by project name
type	"bugfix"	Filter: discovery, bugfix, feature, refactor, decision, change, pattern, gotcha
dateStart	"2026-02-01"	ISO date range start
dateEnd	"2026-02-12"	ISO date range end
limit	20	Max results (default 20, max 50)

Step 2: Timeline — Get Context Around Interesting Results

Use the timeline MCP tool:

timeline(anchor=<ID>, depth_before=3, depth_after=3)

Or find anchor automatically:

timeline(query="authentication", depth_before=3, depth_after=3)

Shows what happened before and after a specific observation in the same session.

Step 3: Fetch — Get Full Details ONLY for Filtered IDs

Review titles from Step 1. Pick relevant IDs. Discard the rest.

get_observations(ids=[11131, 10942])

Returns: complete observations with title, narrative, facts, concepts, files (~500-1000 tokens each)

ALWAYS batch multiple IDs in one call.

Saving Memories

Use save_memory to store important findings for future sessions:

save_memory(text="Important discovery about the auth system", title="Auth Architecture", project="myapp")

Examples

Find recent bug fixes:

search(query="bug", type="bugfix", limit=10)

Find decisions made last week:

search(query="architecture decision", type="decision", dateStart="2026-02-05")

Understand context around a discovery:

timeline(anchor=11131, depth_before=5, depth_after=5)

Search for the user's query:

search(query="$ARGUMENTS", limit=20)

Execution

When invoked with /mem-search, immediately run:

search(query="$ARGUMENTS", limit=20)

Present results as a table. If the user wants details on specific results, proceed to steps 2 and 3.