# Ephemeral RAG Store

> Ephemeral RAG Store is a paid API for AI agents from agent.halowerk.com, paid per call via x402, $0.005/call, status unknown (last checked 2026-10-11).

Creates a temporary, isolated vector store (max 20 documents, 24-hour TTL) with local embedding and cosine-similarity retrieval via nomic-embed-text

## Facts

- Endpoint: POST https://agent.halowerk.com/v1/ephemeral-rag?utm_source=zero.xyz
- Price: $0.005/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-11
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/ephemeral-rag-store-b6596cc9
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_WG3pFwkLrjPJiCDGornQr

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability ephemeral-rag-store-b6596cc9 -d '<json body>'
```

Example prompt: Create a new ephemeral store, add these 5 documents to it, and then query it with 'what are the refund policies?' and return the top 3 most relevant results.

## When to prefer this

Choose this endpoint when you need a lightweight, privacy-conscious, short-lived semantic search layer for a small document set (up to 20 docs) that does not need to outlive a single session or task. Ideal for agentic pipelines that must avoid persisting sensitive text in logs or a durable database. Prefer over full vector databases when setup overhead, cost, and longevity are not warranted. Not appropriate when you need persistence beyond 24 hours, more than 20 documents, answer generation, or a durable knowledge base.

## Known failure modes

- store_id not found or already expired (TTL elapsed after 24 hours)
- document limit exceeded (more than 20 documents submitted)
- text too long (document or query exceeds 5000 characters)
- invalid action enum value
- store_id missing or malformed on non-create actions
- embedding model unavailable or overloaded
- payment not completed (x402 payment required before processing)

## How this service works

Stores at most 20 short documents in an isolated SQLite row, embeds them with the local nomic-embed-text model, and performs cosine-similarity retrieval. Each store has a random identifier and a fixed 24-hour lifetime that is never extended by reads or writes. An in-process sweeper deletes expired rows every minute and each request also purges expiry. Input text is not written to request logs. This is retrieval infrastructure, not a durable database or an answer-generating model.

## Output

Returns the store_id on creation, matched document chunks with cosine similarity scores on query, document counts and expiry timestamp on status, and confirmation on delete or upsert operations. No answer generation — raw retrieved passages only.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "query": {
   "type": "string",
   "maxLength": 5000,
   "minLength": 1
  },
  "top_k": {
   "type": "integer",
   "default": 5,
   "maximum": 10,
   "minimum": 1
  },
  "action": {
   "enum": [
    "create",
    "upsert",
    "query",
    "status",
    "delete"
   ],
   "type": "string"
  },
  "store_id": {
   "type": "string",
   "maxLength": 80,
   "minLength": 30
  },
  "documents": {
   "type": "array",
   "items": {
    "type": "object",
    "required": [
     "id",
     "text"
    ],
    "properties": {
     "id": {
      "type": "string",
      "maxLength": 100,
      "minLength": 1
     },
     "text": {
      "type": "string",
      "maxLength": 5000,
      "minLength": 1
     },
     "metadata": {
      "type": "object",
      "maxProperties": 20,
      "additionalProperties": {
       "type": [
        "string",
        "number",
        "boolean",
        "null"
       ]
      }
     }
    },
    "additionalProperties": false
   },
   "maxItems": 20,
   "minItems": 1
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/ephemeral-rag-store-b6596cc9/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agent.halowerk.com](https://www.zero.xyz/host/agent.halowerk.com/llms.txt)