Launch Alert Bot

Initial Assessment

• Data feeds (Helius token registry events, DEX pool creations, metadata updates)?
• Alert channels and audience? (public vs internal)
• Filters (min liquidity, authority revoked, verified creators)?
• Latency tolerance?

Core Principles

• Prioritize safety: include risk flags and avoid implicit endorsements.
• Deduplicate across feeds; avoid double alerts for same mint/pool.
• Provide verifiable data: addresses, txids, authority states.

Workflow

1. Source selection
   • Monitor pool creation events, new mints with metadata, social signals if desired.
2. Filtering and scoring
   • Apply thresholds: liquidity >= X, metadata exists, mint/freeze authority status, creator whitelist/blacklist.
3. Enrichment
   • Fetch metadata, authority info, initial LP stats; compute simple risk score.
4. Alerting
   • Message with mint/pool addresses, tx link, risk flags (authority retained? low LP? unknown metadata). Include "not financial advice".
5. Operations
   • Deduplicate by mint/pool; throttle; monitor error rates.

Templates / Playbooks

• Alert template: "New pool detected: <mint> on <DEX> at slot <slot>. LP: . Mint authority: revoked/active. Risk flags: ... (tx link)."
• Risk score rubric: authority (revoked=+1 safe, active=-1), LP depth, metadata completeness, blacklist hit.

Common Failure Modes + Debugging

• False positives from testnet/devnet: filter by cluster.
• Spam/scam alerts: strengthen filters, add manual review mode.
• Rate limits on data providers: cache and backoff.
• Metadata fetch failures: retry and mark partial info.

Quality Bar / Validation

• Filters tuned to reduce noise; risk flags included.
• No duplicate alerts for same event.
• Clear disclaimer and source links in every alert.

Output Format

Examples

• Simple: Internal bot watching Raydium pools > liquidity; alerts to Slack with authority status.
• Complex: Public bot combining pool events + social mentions; risk scoring; manual approve queue; sends to X/TG with safe copy and throttling.