Four.Meme Agentic Ops

Usage

• Category: Ecosystem
• Mode: guide
• Version: 0.1.0

Input Example

{
  "goal": "momentum_entry",
  "riskTier": "medium",
  "maxCapitalBnb": 0.6,
  "entrySlices": 3,
  "confirmationMode": "human_in_the_loop"
}

Operating Loop

1. Discover candidates (read-only):

   npx fourmeme token-list
   npx fourmeme token-rankings
   

2. Score each candidate with a fixed rubric:
   • liquidity quality
   • price acceleration
   • holder concentration
   • tax and transfer constraints
3. Build a proposed execution plan (size, slices, and invalidation).
4. Ask for explicit confirmation on every write action.
5. Execute using trade commands from the approved plan:

   npx fourmeme quote-buy ...
   npx fourmeme buy ...
   npx fourmeme quote-sell ...
   npx fourmeme sell ...
   

6. Monitor and journal:

   npx fourmeme events <fromBlock> [toBlock]
   

   Record rationale, execution hash, and post-trade outcome.

Agentic Policy

• Keep default mode as assistive, not autonomous.
• Require explicit user confirmation for all on-chain writes.
• Apply hard risk caps before each order (maxCapitalBnb, maxLossBnb, maxSlippageBps).
• Re-check wallet and quote right before execution to avoid stale state.

Source Anchors

• four-meme-ai package + command list: https://www.npmjs.com/package/four-meme-ai
• Four.meme protocol integration page: https://four-meme.gitbook.io/four.meme/brand/protocol-integration