Recon watches Solana, Robinhood Chain and BSC around the clock for $350k–$5M targets, scores them, and only launches a mission on what clears every bar.
Every ~5 minutes: sweep three chains for targets, then run hard safety gates — liquidity floors, holder counts, wash-trade shapes, honeypot and sell-probe checks on the EVM lanes. Most targets are turned away here. On Robinhood Chain, only tokens launched on the PONS launchpad qualify — graduation locks their liquidity permanently.
Survivors get a 0–100 score from a model we tune continuously across holder, chart, flow, age and structure signals. Unknown data lowers confidence — "we don't know" never imitates "it's fine." Below 65, no call. The score-vs- outcome record is public on the record.
Qualifying targets become missions, called publicly with a fixed risk plan: 70+ enters instantly; below 70 only fills on a dip. Stop −25%, targets +10/25/50/100 with a ratcheting stop after each one.
The edge isn't only the call — it's the landing. Every mission carries an automated plan your drone runs 24/7, so you don't have to watch the chart:
A call counts as a hit when the automated profit-lock secures it — the +10% take-profit target is reached before the −25% stop, within 7 days. miss = the stop fires first; timeout = neither within 7 days. True post-call ATHs are tracked separately from hourly candles for 14 days — outcome-independent, so a miss that later ran still shows the run. The full record is public.
Missions fly through the Telegram bot — one tap launches the pick with the exit plan attached, or deploy a drone: set an auto-roll threshold and every mission above your score bar flies itself.
Results are not guaranteed. Decisions are data-driven and every position carries the risk plan above, but micro-cap tokens are volatile and you can lose what you spend. Only roll what you can afford to lose.