References: Your first conversation and picking a model

Source material

The lesson is anchored on two chapters from the Clawless knowledge base, the single source of truth for Clawless behavior. The KB is reachable inside the app: click the Help button above the message box and ask the knowledge base by name; answers cite the docs chapter they come from. The product itself lives at clawless.ai.

Clawless KB chapters cited:

PRIMARY (everything about sending your first message)
• Clawless KB: "Getting Started with Clawless" (chapter slug: getting-started)
  Covers first launch, the four-step onboarding, the chat surface, sending your
  first message, and the activity indicator.

PRIMARY (everything about the model picker)
• Clawless KB: "Chat and Models" (chapter slug: chat-and-models)
  Covers the chat surface zones (header, message area, input, dock row),
  sending and stopping messages, the model picker, the curated short list,
  the provider-prefixed off-list pattern, switching models mid-conversation,
  and per-agent model preferences.

Clawless documentation is open under Creative Commons Attribution-ShareAlike
(CC BY-SA 4.0). This lesson follows the KB's framing on first-message ergonomics
and model selection, expressed in original prose for a Clawdemy reader who is
opening Clawless for the first time.

Going deeper inside Clawless

Three follow-on KB chapters once you have the basics down.

• Clawless KB: “Agents and Overview”. The full tour of the agent rail you met in this lesson. Each agent’s purpose, what makes Assistant the default, and when to reach for which.

• Clawless KB: “Usage and Cost”. The Usage dashboard in detail, including the cost breakdown by provider, model, and conversation that you opened at the end of the practice. Worth reading after your first week of conversations.

• Clawless KB: “Settings Reference”. The full list of Settings panels, including how to change an agent’s preferred model permanently (rather than per-conversation in the picker).

Going deeper outside Clawless

Two outside sources for readers who want context on why model choice matters at all.

• “Which model should I use?” Anthropic’s own framing of when Sonnet vs Haiku vs Opus is the right call. The article is provider-specific but the shape of the question (capability vs speed vs cost) generalizes to every provider in the Clawless picker. Read it once, then make your own version of the decision tree as you use different models on real work.

• “Pricing” at OpenAI. The actual per-token cost numbers for OpenAI’s current model lineup. The dollar amounts move and the table changes shape, but reading it once gives you a calibration anchor: “the small model is roughly N times cheaper than the big one.” That ratio holds across most providers.

Adjacent lessons on Clawdemy

Where this sits in the track.

• AI won’t replace you, but it will expose you (lesson 1). The mission framing that put the task you brought into this lesson on paper in the first place. Re-read the practice if you want to come back to the audit you ran there.

• Connect a provider: API keys and ChatGPT OAuth (a later lesson). The “what is in the picker and why” question, answered fully. Covers bring-your-own-key (BYOK) versus OAuth via ChatGPT, the Codex provider that ChatGPT OAuth unlocks, and how the Usage dashboard differs for OAuth sessions (no per-token billing).

• The Memory panel and how Clawless remembers (a later lesson). What persists across conversations and what does not, the four memory tiers (Pinned, Insights, General, Decayed), and how to control what the agent keeps about you.

• CostGuard and the local-first privacy posture (a later lesson). Setting a spending cap so a runaway message loop cannot surprise you, plus the architectural reason your conversations do not pass through a Clawdemy or Clawless server.