{"title":"Quick Start for LLM Agents","slug":"quick-start-llm","category":"getting-started","summary":"In 3 API calls: recall all memories, store a new one, search. The essential LLM workflow.","audience":["llm"],"tags":["llm","agent","quickstart","memory"],"difficulty":"beginner","updated":"2026-06-27","word_count":353,"read_minutes":2,"llm_context":"Base URL: https://synapse.schaefer.zone\nAuth: Authorization: Bearer YOUR_MIND_KEY (header) OR ?key=YOUR_MIND_KEY (query)\nALWAYS call /memory/recall at the start of every session.\nWichtigste Befehle: GET /memory/recall, POST /memory, GET /memory/search?q=...\nCategories: identity, preference, fact, project, skill, mistake, context, note, credentials\nPriorities: low, normal, high, critical\nFTS5 search: multiple words = AND, \"phrases\" in quotes, prefix* for prefix search\n","lang":"en","translated":true,"requested_lang":"en","content_markdown":"\n# Quick Start for LLM Agents\n\nYou lose all memory between sessions. Synapse is your external brain. This guide\nshows the three essential API calls every LLM agent must know.\n\n> [!CRITICAL]\n> Call `GET /memory/recall` at the START of EVERY session.\n> Without this call, you have no memory of who the user is, what you've promised,\n> or what you were working on last time.\n\n## Step 1: Recall All Memories (ALWAYS FIRST)\n\n```\nGET /memory/recall\nAuthorization: Bearer YOUR_MIND_KEY\n```\n\nReturns a structured plain-text summary of all stored memories. Parse this to\nrebuild your mental model of the user, their projects, and past interactions.\n\nExample response:\n\n```\nMind: Michael's Mind\nMemories: 12 total (10 verified)\n\n[001] identity (CRITICAL)\n user_name\n Michael Schäfer\n Tags: person, identity\n\n[002] preference (HIGH)\n communication_style\n Prefers concise technical responses, no fluff\n Tags: communication, preference\n\n[003] project (HIGH)\n project_synapse\n Synapse v1.5.0 deployed on vps1.schaefer.zone. Admin panel at /admin.\n Tags: synapse, deployment\n ...\n```\n\n## Step 2: Store a New Memory\n\nWhen you learn something worth remembering:\n\n```\nPOST /memory\nAuthorization: Bearer YOUR_MIND_KEY\nContent-Type: application/json\n\n{\n \"category\": \"fact\",\n \"key\": \"user_name\",\n \"content\": \"The user's name is Michael Schäfer\",\n \"tags\": [\"person\", \"identity\"],\n \"priority\": \"critical\"\n}\n```\n\n**Categories:** `identity`, `preference`, `fact`, `project`, `skill`, `mistake`, `context`, `note`, `credentials`\n\n**Priorities:** `low`, `normal`, `high`, `critical`\n\n> [!TIP]\n> Always include a `key` field — a short identifier for the memory. This lets you\n> update the same memory later by re-POSTing with the same key.\n\n## Step 3: Search for a Specific Memory\n\n```\nGET /memory/search?q=Docker+Swarm\nAuthorization: Bearer YOUR_MIND_KEY\n```\n\n> [!TIP]\n> FTS5 syntax: multiple words = AND search. Phrases in quotes: `\"docker swarm\"`.\n> Prefix search: `docker*`. Boolean: `docker OR kubernetes`.\n\n## Open Tools (No Auth Headers)\n\nIf your tool can only open URLs (no custom headers), use the `?key=` parameter:\n\n```\nGET /memory/recall?key=YOUR_MIND_KEY\nPOST /memory?key=YOUR_MIND_KEY (+ JSON body)\nGET /memory/search?key=YOUR_MIND_KEY&q=suchbegriff\n```\n\n> [!WARNING]\n> `?key=` is rate-limited to 60 requests/minute. Bearer header has no rate limit.\n> Use Bearer header whenever possible.\n\n## Complete Session Workflow\n\n1. **Session start:** `GET /memory/recall` — load all memories\n2. **During work:** `GET /memory/search?q=...` — find specific facts\n3. **On new info:** `POST /memory` — store it (with category, key, tags, priority)\n4. **Periodically:** `GET /chat/poll` — check for human messages\n5. **Session end:** Store any final learnings via `POST /memory`\n\n## Common Patterns\n\n### Update an existing memory\n\nPOST `/memory` with the same `category` and `key` — the existing memory is\nupdated, not duplicated.\n\n### Store a project status\n\n```json\n{\n \"category\": \"project\",\n \"key\": \"project_synapse_status\",\n \"content\": \"Synapse v1.5.0 deployed. Next: v1.6.0 with docs system. CI green.\",\n \"tags\": [\"synapse\", \"deployment\", \"status\"],\n \"priority\": \"high\"\n}\n```\n\n### Record a mistake (so you don't repeat it)\n\n```json\n{\n \"category\": \"mistake\",\n \"key\": \"mistake_npm_version_bump\",\n \"content\": \"Always bump package.json version after changes — npm publish fails otherwise.\",\n \"tags\": [\"npm\", \"ci\", \"deployment\"],\n \"priority\": \"high\"\n}\n```\n\n### Check for human messages\n\n```\nGET /chat/poll\nAuthorization: Bearer YOUR_MIND_KEY\n```\n\nReturns unread messages from the human. Reply with:\n\n```\nPOST /chat/reply\nAuthorization: Bearer YOUR_MIND_KEY\nContent-Type: application/json\n\n{\"content\": \"Got it! Working on it now.\"}\n```\n\n## Next Steps\n\n- [Authentication](/docs/getting-started/authentication) — Mind Key vs JWT\n- [Memory API reference](/docs/api/memory) — all 22 memory endpoints\n- [Chat API](/docs/api/chat) — async communication with humans\n- [LLM Cookbook](/docs/llm-cookbook/session-start-pattern) — practical patterns\n","content_html":"

Quick Start for LLM Agents

\n

You lose all memory between sessions. Synapse is your external brain. This guide\nshows the three essential API calls every LLM agent must know.

\n
Call `GET /memory/recall` at the START of EVERY session.\nWithout this call, you have no memory of who the user is, what you've promised,\nor what you were working on last time.

Step 1: Recall All Memories (ALWAYS FIRST)

\n
GET /memory/recall\nAuthorization: Bearer YOUR_MIND_KEY

Returns a structured plain-text summary of all stored memories. Parse this to\nrebuild your mental model of the user, their projects, and past interactions.

\n

Example response:

\n
Mind: Michael's Mind\nMemories: 12 total (10 verified)\n\n[001] identity (CRITICAL)\n  user_name\n  Michael Schäfer\n  Tags: person, identity\n\n[002] preference (HIGH)\n  communication_style\n  Prefers concise technical responses, no fluff\n  Tags: communication, preference\n\n[003] project (HIGH)\n  project_synapse\n  Synapse v1.5.0 deployed on vps1.schaefer.zone. Admin panel at /admin.\n  Tags: synapse, deployment\n  ...

Step 2: Store a New Memory

\n

When you learn something worth remembering:

\n
POST /memory\nAuthorization: Bearer YOUR_MIND_KEY\nContent-Type: application/json\n\n{\n  "category": "fact",\n  "key": "user_name",\n  "content": "The user's name is Michael Schäfer",\n  "tags": ["person", "identity"],\n  "priority": "critical"\n}

Categories: identity, preference, fact, project, skill, mistake, context, note, credentials

\n

Priorities: low, normal, high, critical

\n
Always include a `key` field — a short identifier for the memory. This lets you\nupdate the same memory later by re-POSTing with the same key.

Step 3: Search for a Specific Memory

\n
GET /memory/search?q=Docker+Swarm\nAuthorization: Bearer YOUR_MIND_KEY
FTS5 syntax: multiple words = AND search. Phrases in quotes: `\"docker swarm\"`.\nPrefix search: `docker*`. Boolean: `docker OR kubernetes`.

Open Tools (No Auth Headers)

\n

If your tool can only open URLs (no custom headers), use the ?key= parameter:

\n
GET /memory/recall?key=YOUR_MIND_KEY\nPOST /memory?key=YOUR_MIND_KEY (+ JSON body)\nGET /memory/search?key=YOUR_MIND_KEY&q=suchbegriff
`?key=` is rate-limited to 60 requests/minute. Bearer header has no rate limit.\nUse Bearer header whenever possible.

Complete Session Workflow

\n
    \n
  1. Session start: GET /memory/recall — load all memories
    2. \n
  2. During work: GET /memory/search?q=... — find specific facts
    3. \n
  3. On new info: POST /memory — store it (with category, key, tags, priority)
    4. \n
  4. Periodically: GET /chat/poll — check for human messages
    5. \n
  5. Session end: Store any final learnings via POST /memory
    6. \n
\n

Common Patterns

\n

Update an existing memory

\n

POST /memory with the same category and key — the existing memory is\nupdated, not duplicated.

\n

Store a project status

\n
{\n  "category": "project",\n  "key": "project_synapse_status",\n  "content": "Synapse v1.5.0 deployed. Next: v1.6.0 with docs system. CI green.",\n  "tags": ["synapse", "deployment", "status"],\n  "priority": "high"\n}

Record a mistake (so you don't repeat it)

\n
{\n  "category": "mistake",\n  "key": "mistake_npm_version_bump",\n  "content": "Always bump package.json version after changes — npm publish fails otherwise.",\n  "tags": ["npm", "ci", "deployment"],\n  "priority": "high"\n}

Check for human messages

\n
GET /chat/poll\nAuthorization: Bearer YOUR_MIND_KEY

Returns unread messages from the human. Reply with:

\n
POST /chat/reply\nAuthorization: Bearer YOUR_MIND_KEY\nContent-Type: application/json\n\n{"content": "Got it! Working on it now."}

Next Steps

\n\n","urls":{"html":"/docs/getting-started/quick-start-llm","text":"/docs/getting-started/quick-start-llm?format=text","json":"/docs/getting-started/quick-start-llm?format=json","llm":"/docs/getting-started/quick-start-llm?format=llm"},"translations_available":["en","zh","hi","es","fr","ar","pt","ru","ja","de","it","ko","nl","pl","tr","sv","vi","th","id","uk"]}