Stats
Actions
Tags
From quant-llm-skills
Use when an SEC filing names a placement agent, underwriter, or sales agent (e.g., Goldman Sachs, H.C. Wainwright, Maxim, Aegis, Roth, B. Riley, Cantor, Jefferies). Classifies the firm into a 4-tier framework that materially changes the dilution risk profile of the offering.
How this skill is triggered — by the user, by Claude, or both
Slash command
/quant-llm-skills:bank-tier-classificationThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
The same filing (424B5, 8-K) means very different things depending on
The same filing (424B5, 8-K) means very different things depending on who is selling the shares. A registered direct led by Goldman Sachs is a different animal than an ATM run by a tier-4 specialist. Most quant pipelines ignore the agent identity entirely; this skill restores it as a primary signal.
The placement agent is a leading indicator of the offering's structure, aggression, and post-deal price behavior.
Tier 1 firms protect their reputation and underwrite institutional demand. Tier 4 firms specialize in placing offerings that tier 1 firms will not touch — by definition, those are the harder, more dilutive, more structured deals.
Same form, different agent = different trade.
Global investment banks underwriting at scale. Their participation implies institutional demand and a "blessed" issuer.
Implication when seen on a small-cap filing: rare. If present, the issuer is a real institutional name or has a strategic reason for the relationship. Dilution risk: low to moderate, deal is typically clean.
Quality investment banks with research coverage and institutional distribution. Do legitimate growth-capital raises.
Implication: legitimate growth capital. Often firm-commitment underwritten secondaries, not ATMs. Dilution risk: moderate, predictable.
Reputable smaller banks with niche specialties (biotech, energy, tech). Mix of clean and aggressive deals depending on the issuer.
Implication: mixed bag — read the structure (warrants, discount, ATM language) carefully. The agent identity alone is not decisive at this tier. Dilution risk: moderate to high depending on deal terms.
Firms that specialize in placing offerings for issuers that cannot attract Tier 1–3 underwriting. Heavy concentration in ATMs, registered directs with warrants, ELOCs, and PIPEs. High-volume, high-frequency small-cap deal flow.
Implication: when one of these agents appears on a 424B5 or sales agreement, the base rate for: (a) attached warrants, (b) deep discount to last trade, (c) ATM that sells aggressively into intraday strength, (d) repeat dilution within 30–90 days, is materially elevated. Dilution risk: HIGH. Treat the issuer as having a continuous selling pressure floor until the offering is exhausted or terminated.
"Is this offering institutional or retail-flow?" Tier 1–2 = institutional book. Tier 4 = retail / quant / market-maker absorption with frequent shorting against the cross.
"How fast will this dilute?" Tier 4 ATMs sell every green tick. Tier 2 underwritten secondaries are one-shot. The agent tier is the strongest predictor of cadence.
"Should I go long after this offering closes?" Tier 1–2 underwritten secondaries often see post-deal stabilization (the underwriter has incentive to support the price). Tier 4 ATMs have the opposite incentive — keep selling.
"Multiple agents listed — which tier wins?" Use the LOWEST (most aggressive) tier present. A "Tier 2 + Tier 4" joint placement is dominated by the tier-4 dynamics.
atm-detection if the structure is an ATM,
and lookahead-safety for any historical timing claims.This is not a firm-rating service. It is a heuristic that maps observed
firm names to expected deal behavior based on documented patterns of
small-cap deal flow. Always combine with structure (atm-detection)
and timing (lookahead-safety) for a complete read.
npx claudepluginhub jefrnc/quant-llm-skills --plugin quant-llm-skillsSearches MemPalace before answering questions about past work, people, projects, or prior decisions. Returns verbatim stored content instead of guessing from model memory.
Guides Payload CMS config (payload.config.ts), collections, fields, hooks, access control, APIs. Debugs validation errors, security, relationships, queries, transactions, hook behavior.
Implements vector databases with Pinecone, Weaviate, Qdrant, Milvus, pgvector for semantic search, RAG, recommendations, and similarity systems. Optimizes embeddings, indexing, and hybrid search.