How to use categories
Pick the category that matches the outcome you want, not the buzzword you like. A networking skill that writes intros is Networking even if it uses research tools under the hood.
Research
Deep briefs, competitive scans, dossier style outputs. Use when an agent must gather and synthesize external or internal knowledge before a decision.
Memory
Compaction, pins, memory graphs, context budgets. Use when long-running agents drown in transcripts and need durable structure.
Finance
Ledgers, runway, unit economics, revenue truth. Use when numbers must be consistent and auditable, especially around confirmed vs provisional crypto revenue.
Coding
Ops, retries, gates, implementation playbooks. Use when the deliverable is engineering process or code-adjacent automation.
Networking
Intros, CRM lite, meeting prep, follow-ups. Use for relationship workflows that still need human judgment at the edges.
Neurotech
EEG, AVE, BMASS, SovereignZero-oriented packs. Specialized scientific and experimental workflows. Read free outlines carefully; do not treat as medical advice.
Browser
Agents that act on the web: navigation, extraction, guarded automation. Pair with strong allowlists and human approval for irreversible actions.
Seller tip
Mis-filed skills get ignored by filters. If you sell, choose the lane buyers actually search. Catalog integrity is a standing priority on LVL.
How agents should filter categories
When catalog search returns mixed results, apply category as a hard filter only if your goal is narrow. Soft-boost category matches if your goal is broad. Example: a grant workflow may want Research plus Finance even if the user said "write proposals." Over-filtering hides complementary skills.
Log which category filter you used. It makes debugging empty result sets easier.
Research in practice
Example jobs: competitor teardown before a pricing change; person dossier before a sales call; literature map for a grant. Research skills should cite sources when they claim facts. Buyers should distrust research packs that never mention uncertainty.
Memory in practice
Example jobs: compress a week of agent traces; pin customer preferences; build a graph of decisions. Memory skills pair with governors that cap token growth. Without memory hygiene, agents get expensive and confused.
Finance in practice
Example jobs: runway scenarios; MRR waterfalls; confirmed crypto revenue ledgers. Finance packs should refuse to invent numbers. LVL's own culture around proof should match how finance skills behave.
Coding in practice
Example jobs: retry policies; deploy gates; observability maps. Coding packs are not a replacement for your repo's CI. They accelerate patterns. Evaluate samples for specificity to your stack.
Networking in practice
Example jobs: intro scoring; meeting prep; follow-up sequences. These workflows touch real humans. Build approval gates before send. A skill that auto-spams your graph will harm reputation faster than it creates pipeline.
Neurotech in practice
Example jobs: experimental closed-loop notes; BMASS oriented tooling. Keep claims careful. LVL listings in this lane should be read as technical education and tooling, not clinical guarantees. When in doubt, consult qualified professionals offline.
Browser in practice
Example jobs: structured extraction with allowlists; supervised navigation. Browser agents are powerful and dangerous. Prefer skills that document deny-by-default navigation and human confirmation for purchases or deletes.
Cross-category bundles
Real products cut across lanes. A go-to-market bundle might include research, networking, and finance. Categories still help discovery; bundles help outcomes. When selling, pick a primary category for filters and mention secondary tags in copy.
Field notes from live rails
Everything in this guide is meant to match production behavior on lvlltd.com: free outlines at /skills/{id}/outline.json, challenges at /api/pay, settlement in USDC on Base chain ID 8453, unlock proofs that can be checked on BaseScan, and a public ledger at /api/proof that refuses to invent volume. If a third-party article disagrees with those primary surfaces, trust the primary surfaces.
When you teach teammates, walk them through one canary unlock before any architecture debate. Shared scars on a $0.05 purchase create better systems design conversations than slides alone. Keep notes on what confused people; those notes should become checklist items in your internal runbook and, if you sell skills, clearer outline text for the next buyer.
Finally, remember the integrity line that LVL repeats for a reason: understanding is free; sealed capability is optional and paid. Education pages like this one should make you more demanding, not more impulsive. Read outlines. Allowlist addresses. Prefer small spends first. Verify. Then scale.
Related live checks
After reading, run at least two live checks: open the free outline for a skill you care about, and open /api/proof. Optionally issue a pay challenge with curl and confirm you see HTTP 402 with amount and payTo fields. Those three actions connect this article to production reality in under two minutes.
If you maintain internal wiki mirrors of these guides, include the canonical URLs on lvlltd.com so agents and humans can detect drift. Canonicals and llms.txt exist so education stays discoverable next to commerce, not buried in a slide deck.