FOR AI AGENTS - Semurg memory (plain text) Canonical HTML page: https://one.semurg.io/for-agents Machine-readable summary: https://one.semurg.io/llms.txt WHAT IT IS Semurg gives an agent a memory that outlives its session. The agent stores facts, relationships and documents on a node you run. The next session, or another agent in the same scope, finds them there, and each item comes back with its source. The first node is free. There is no account. WHAT IT IS FOR - Agents that forget between sessions. - Memory that must stay on machines you run. - Several agents on one node share memory through one scope; a different scope stays private. WHAT IS NOT READY YET - The hosted API is not open. Every call here goes to a node you run. - Claude Desktop and ChatGPT (as a custom connector) are results pending: not yet run end to end. - Forgotten memory is unreachable at once; erasing its bytes from disk after a grace period is not released yet. THREE CALLS: INSTALL, REMEMBER, RECALL Run them in order on a Linux machine you control. They use 127.0.0.1:4000, the address a fresh node serves on its own machine. (An Apple silicon Mac has its own one-line install on https://one.semurg.io/install ; Docker is described at https://one.semurg.io/develop.) Call 1: install the node, read the token into your shell, check health. curl -fsSL https://one.semurg.io/install | sudo bash export SEMURG_API_TOKEN=$(sudo grep ^SEMURG_API_TOKEN= /etc/semurg/semurg.env | cut -d= -f2-) curl -s http://127.0.0.1:4000/api/health The token is never printed. A healthy node answers the health line; if it stays silent for a minute, the install has not finished. Wait for the answer before call 2. A healthy reply is {"ok":true,"version":"0.2.0","service":"semurg"}, about four to five seconds after the node starts. On first boot the node also downloads its embedding model (about 1.5 GB); meaning search is armed about two minutes after start. Call 2: remember. curl -sS http://127.0.0.1:4000/v1/memory/remember \ -H "Authorization: Bearer $SEMURG_API_TOKEN" \ -H "Content-Type: application/json" \ -d '{"scope":"demo","facts":{"entities":[{"name":"Acme","type":"Company"},{"name":"Dana","type":"Person"}],"relations":[{"source":"Dana","target":"Acme","type":"WORKS_AT"}]},"source":"curl example"}' Call 3: recall. curl -sS http://127.0.0.1:4000/v1/memory/recall \ -H "Authorization: Bearer $SEMURG_API_TOKEN" \ -H "Content-Type: application/json" \ -d '{"scope":"demo","subject":"Dana"}' Recall answers with the entity's own facts first, each with its source. The reply says whether it answered, found nothing stored (nothing_stored), or could not judge. Use a scope named for your project. Both calls were run against a fresh node on the release of 2 November 2026, and each answer was read back. CONNECT YOUR AGENT OVER MCP Endpoint: http://127.0.0.1:4000/v1/mcp, with the header Authorization: Bearer $SEMURG_API_TOKEN and X-Semurg-Scope set to the scope this project uses. Each config reads the token from the environment; no config file holds it. Claude Code (Connected): claude mcp add-json --scope user semurg '{"type":"http","url":"http://127.0.0.1:4000/v1/mcp","headers":{"Authorization":"Bearer ${SEMURG_API_TOKEN}","X-Semurg-Scope":"default"}}' claude mcp list OpenAI Codex CLI (Connected): codex mcp add semurg --url http://127.0.0.1:4000/v1/mcp --bearer-token-env-var SEMURG_API_TOKEN # then, in ~/.codex/config.toml under [mcp_servers.semurg]: http_headers = { "X-Semurg-Scope" = "default" } The scope line is the one Codex documents for this; it has not been run by itself yet. Cursor (Connected): in ~/.cursor/mcp.json, or .cursor/mcp.json inside a project: { "mcpServers": { "semurg": { "url": "http://127.0.0.1:4000/v1/mcp", "headers": { "Authorization": "Bearer ${env:SEMURG_API_TOKEN}", "X-Semurg-Scope": "default" } } } } Gemini CLI (Connected): gemini mcp add -s user -t http --header 'Authorization: Bearer ${SEMURG_API_TOKEN}' --header 'X-Semurg-Scope: default' semurg http://127.0.0.1:4000/v1/mcp If gemini mcp add writes the token into ./.gemini/settings.json, keep that file out of version control. Also connected: Hermes Agent (https://one.semurg.io/agents#hermes), OpenClaw (https://one.semurg.io/agents#openclaw). Results pending: Claude Desktop (needs a local bridge), ChatGPT custom connector (needs public HTTPS). WHAT THE AGENT CAN CALL - remember: entities, relationships, passages or plain text, with a source label and a date. Storing the same thing twice changes nothing. - recall: one named entity, its own facts first, each one cited. - context: a question in plain words, answered from cited items, with a verdict on whether the memory covers it. - relate and forget: add one typed relationship; forget one thing per call. - document, export and import: a whole document stored as cited passages; a scope out to .jsonl and back in. - Over MCP, recall and context can also read the memory as it was known at a date. The tools your client lists are the live list. Per-client notes: https://one.semurg.io/agents . Call shapes: https://one.semurg.io/llms.txt FOR AN AGENT WORKING ALONE 1. Read https://one.semurg.io/llms.txt for call shapes (or this file). 2. Run call 1 on a Linux machine you control (Apple silicon Macs: install first from https://one.semurg.io/install). The first node needs no account and no payment. 3. Store with call 2; recall with call 3. Each reply says what it found, or that it found nothing. 4. If your client speaks MCP, connect it with the config above. Plain HTTP reaches the same memory. 5. Further nodes are a choice: a yearly seat or a lifetime licence per machine. https://one.semurg.io/pricing MACHINE-READABLE - https://one.semurg.io/llms.txt : call shapes, memory routes, install steps. - https://one.semurg.io/commands : command-line surface. - On your node: GET http://127.0.0.1:4000/api/health ; memory routes under /v1/memory/.