memory_searchLABSCO SUMMARY
Due to be replaced, and explicitly disposable.
The maintainer is unusually direct about it: the current tool surface is being cut in full, with no data import into what replaces it. That is worth knowing before you build a workflow on memory_record. The source repository is no longer publicly accessible, and the maintainer has announced that this toolset will be replaced: run it only if the cross-session recall is worth that, and keep anything you would miss somewhere outside it.
What it is
A Go memory service for agents, running as a container with its own vector store. The shipped build records strategies that worked, error patterns and their fixes, and session checkpoints, and searches them back at the start of the next task. The README announces a complete rewrite aimed at knowledge shared between engineering teams and their agents — Incidents, Decisions and Standards attached to an Org, Team or Project — with no migration and no compatibility shim from what runs today.
What you get
- Memory:
memory_search,memory_record,memory_feedback,memory_outcome,memory_consolidateandmemory_consolidate_session— memories carry an outcome of success or failure and a confidence score that moves with feedback - Checkpoints:
checkpoint_save,checkpoint_listandcheckpoint_resumefor picking a long task back up - Error patterns:
remediation_search,remediation_recordandremediation_feedback - Sub-task isolation with token budgets:
branch_create,branch_return,branch_status - Code and conversation indexing:
repository_index,repository_search,semantic_search,conversation_index,conversation_search - Diagnostics:
troubleshoot_diagnose,reflect_report,reflect_analyze
Requirements
The container ghcr.io/fyrsmithlabs/contextd:latest, run over stdio with a named volume mounted at /data. Embeddings run locally through fastembed, so no cloud key is needed; the vector store is embedded by default and Qdrant is optional, configured with QDRANT_HOST and QDRANT_PORT. CONTEXTD_DATA_PATH defaults to /data.
Setup effort
Run a container — pull the image and point your client at it
USE CASES
01
Not solving the same bug twice
Record the error signature and the fix that worked, then search that record the next time the same failure appears.
02
Starting a task with what worked last time
Search past sessions for the approach you used on a similar problem before the agent goes exploring.
03
Resuming a long job after a restart
Save a checkpoint at a natural break and load it back rather than re-explaining the state.
04
Searching your own past conversations
Index a Claude Code history and query it as a corpus instead of scrolling back through transcripts.
23 tools
Listed in project docs. Your client may show a different list.
memory_recordmemory_feedbackmemory_outcomememory_consolidatememory_consolidate_sessioncheckpoint_savecheckpoint_listcheckpoint_resumeremediation_searchremediation_recordremediation_feedbackbranch_createbranch_returnbranch_statusrepository_indexrepository_searchsemantic_searchconversation_indexconversation_searchtroubleshoot_diagnosereflect_reportreflect_analyzeBEFORE YOU INSTALL
What to know first
Tools you get
23 tools named in the project's documentation.
Requires
docker on PATH
Where it runs
Both documented clients run the server on your machine.
INSTALLATION
Connect your client
MCP Clients
Claude Code, Claude Desktop
JSON
{
"mcpServers": {
"contextd": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-v",
"contextd-data:/data",
"-e",
"CONTEXTD_EMBEDDINGS_PROVIDER=fastembed",
"ghcr.io/fyrsmithlabs/contextd:latest",
"--mcp",
"--no-http"
]
}
}
}
