Choosing the Right SPRAPP Tool for the Job
A practical decision guide for when to reach for SPRAPP Panel, smoltext, TinyLM, or Sprappy Filter — and how they fit together across industries.
Four Tools, Four Jobs
The SPRAPP suite is most powerful when each product is used for what it is good at, rather than forced to do everything. This guide maps common needs to the right tool, drawing on patterns that recur across legal, healthcare, finance, devops, support, and more.
Reach for SPRAPP Panel When the Decision Matters
Use SPRAPP Panel when being wrong is costly and the answer is ambiguous. You ask once, a panel of models reasons independently and converges, and disagreement among them is a feature — it tells you exactly where to apply human judgment. Examples: contract clause analysis, borderline fraud cases, ambiguous compliance flags, postmortem root-cause reasoning.
Reach for Sprappy Filter at Every Boundary
Use Sprappy Filter wherever untrusted content enters your system. It scores inputs across 25 threat categories before any reasoning model sees them, catching prompt injection and adversarial payloads. Filtering first means everything downstream only reasons over content that has cleared a threat check. This applies to support inboxes, RAG retrieval output, public-records portals, and review intake alike.
Reach for smoltext When Short Strings Pile Up
Use smoltext when you are storing or moving many short strings — IDs, codes, tags, status labels, log fields. General compressors like gzip waste bytes on small-payload overhead; smoltext is built for exactly this. The savings show up in audit logs, ticket stores, catalog metadata, and citation databases.
Reach for TinyLM When You Need Offline or Cheap
Use TinyLM when work happens off-network, in privacy-restricted settings, or at a volume where per-call cloud cost hurts. Its eeny and meeny models run entirely on-device, enabling offline drafting in the field, local-first intake of sensitive data, and effectively free high-volume drafting.
The Suite as a System
The strongest workflows combine them: Filter guards the input, Panel makes the hard decision, smoltext shrinks the resulting logs, and TinyLM handles offline or high-volume drafting. Each tool stays in its lane, and the human stays accountable.
A Simple Starting Rule
Begin with Filter at your most exposed input, add the one other tool that matches your biggest pain — cost, offline, storage, or decision quality — and expand from there.