AI DevelopmentLenny's Newsletter

What a harness is and how to build one with Claude Agent SDK

Read original
ai-coding-toolsautomation-stackspkm-workflows

“It's not the model, it's the harness—a purpose-built agent system with encoded permissions, structured inputs, and opinionated integrations that eliminates repetitive prompting”

Key takeaways

  • A 'harness' is a purpose-built agent wrapper with three core components: structured input handling, permission-scoped tool access (Sentry, Linear, GitHub, Vercel), and opinionated output formatting—not a general-purpose chatbot
  • The harness pattern eliminates repetitive natural language prompting by encoding domain logic, permissions, and workflows into the agent architecture itself (e.g., 'fix this bug' becomes automated evidence gathering → root-cause analysis → artifact creation)
  • Claude Agent SDK + custom terminal UI (Ink library) + opinionated adapters create a replicable template for building domain-specific agents; GPT-5.5 and Claude Opus both initially resisted the architecture pattern, suggesting this is non-obvious design
  • Harnesses are most valuable for repetitive, structured workflows with clear inputs/outputs and bounded tool access—not for open-ended tasks requiring general reasoning

Why this matters for operators: Engineering teams evaluating AI agent frameworks; product builders considering agent-first architecture; DevOps/platform teams automating structured workflows

I cover AI×GTM intelligence like this every Wednesday.

Get STEEPWORKS Weekly

More picks

AI DevelopmentLenny's Podcast

Humans will keep inventing new reasons why we must stay in the loop with agents

  • Human resistance to full AI autonomy is not purely technical—it's psychological and organizational; companies will rationalize keeping humans in decision loops even when agents are capable
  • The 'human-in-the-loop' requirement may become a self-perpetuating narrative rather than a genuine necessity, driven by organizational risk aversion and change resistance
  • Product leaders at scale (Notion) are observing this pattern, suggesting it's a widespread phenomenon across enterprise AI adoption, not isolated to specific use cases
ai-agent-adoptionhuman-in-the-loopai-governance
Read original Full analysis →
GTM Ops**RevOps Impact (Jeff Ignacio)

Comp plans for consumption pricing

  • Consumption pricing fundamentally breaks traditional SaaS comp models—requires rethinking sales incentive structures around usage vs. contract value
  • Four distinct contract structures exist (pay-as-you-go, uncommitted, committed, hybrid), each requiring different compensation mechanics and sales behaviors
  • Enterprise consumption-based deals create tension: customers want flexibility, sales teams need predictability for quota attainment—comp design must bridge this gap
revenue-platform-consolidationconsumption-pricing-modelssales-comp-design
Read original Full analysis →
AI×GTMGTM OS: The Future GTM Operator

3 revenue motions your AI is only half wired into

  • Model parity has arrived: OpenAI/Claude now trade evenly on core tasks, making 'better AI' a non-differentiator—the edge shifts to integration depth into existing revenue motions
  • Waste is quantified: teams paying $17K-$37K/month for AI seats that never touch pipeline generation; real cost is opportunity cost of unused capacity, not subscription fees
  • Lean teams have a structural advantage: cannot out-buy larger competitors on model access, but can out-embed them by wiring AI 1 revenue motion deep (pipeline → content → deals) with proprietary deal context competitors haven't seen
ai-sdr-adoptionrevenue-platform-consolidationback-to-basics-gtm
Read original Full analysis →

This analysis was produced using the STEEPWORKS system — the same agents, skills, and knowledge architecture available in the GrowthOS package.