Wrappers are prompts, harnesses are bodies: a VC's thesis on B2B agent infrastructure
MartinGTobias · x · 2026-11-02
Martin Tobias of Incisive Ventures lays out his thesis: a wrapper is essentially a complex prompt, thin relative to what the LLM contributes. A harness gives the brain a body — the job (workflow, proprietary data, integrations, guardrails) plus the scaffolding (tools, memory, evals, orchestration) that let an agent finish multi-step work. Two investable properties: model-agnostic (swap Claude for GPT/Gemini and the company still works) and owns the customer relationship. Both leading labs use the term: Anthropic describes the harness around Claude (progress logs, git checkpoints, end-to-end testing), and OpenAI calls the Codex agent loop its "harness". Context: the pervasive fear that every OpenAI feature launch kills a hundred startups overnight.
More from coding & agent
- Vibe coding's 5-year leap: from a GPT-3 React button demo to an Opus 5.5 three.js world — bennash · 2026-11-03
- Stripe's Link Agent Wallet lands in Merge Agent Handler: agents pay without seeing card details — shensi · 2026-11-03
- Ask Your Agent to Reverse-Engineer HTTP Calls Into a CLI, Skip Slow Computer-Use — vivekhaldar · 2026-11-03
- The 'Illustrated LLM Book' repo is free: 12 chapters of runnable Colab notebooks — mdancho84 · 2026-11-03
- Google open-sources Elephant-Goldfish Model: agent skills for software design and review — adnan_hashmi · 2026-11-03
- CircleCI report: work in progress is the new software delivery bottleneck — rseroter · 2026-11-03