conviction-funnellisted

# conviction-funnel Turn a market full of noise into a *small* set of researched, actionable names. The premise: any single scan can fire on a fluke, and a momentum name is usually already extended by the time it ranks — so picking off one screener tends to buy tops. This funnel stacks **independent** signals (trend, basing, oversold-bounce, options flow) so that only names where 2–3 of them *agree* survive, then spends real research effort only on that handful — ending in a side-by-side table of where to enter, where the stop goes, how big to size, and what kills the thesis. It orchestrates skills the user already has rather than re-implementing anything: ``` regime-scan ── market gate: 🟢/🟡/🔴 + divergence flags (are we even adding risk?) momentum-scan ── primary name list + per-name buyability (Sig), stops, persistence cross-scan ── which names show up in 2+ of the four sister scans (the convergence) │ ▼ select top-N by a risk/reward lens yfinance + edgartools + web + (conditional) wallstreetbets │ ▼ per-name entry / stop / size / invalidation, regime threaded into sizing ``` Default N is **3**. The user can ask for more ("give me 5") or fewer. `<SKILLS_DIR>` below is the parent directory holding the sister-scan folders. Default to `~/.claude/skills` (the individual scan folders there are symlinks into the user's skills repo, so reading/writing state is consistent). If the scans aren't found there, fall back to the repo those symlinks point at — f