ThoughtsClaimsProjectsPredictionsChanged my mindDecisionsQuestionsPostsInputsObservationsAboutPrivacyGraphInteractive 3D rendering of the knowledge network. Drag to rotate, scroll to zoom, and select a node to open it.Knowledge graph indexThe interactive 3D graph requires JavaScript. Below is the same data as a plain list.ClaimsAgent harnesses absorb complexity that would otherwise require separate infrastructure scaffolding such as service meshes or orchestration engines.An agent runtime allows a small set of well-written instructions to substitute for what previously required a small platform.Agent-based systems collapse the surface area of complexity: the branching is real but requires no extra system to deploy, monitor, version, or onboard people to.Shifting cost from infrastructure to writing well is a favorable trade-off.Decision logic that previously required state machines, config files, DSLs, or feature-flag services can now be expressed as a few paragraphs of natural-language instruction inside an agent harness.In an agent-based system, the prompt simultaneously serves as the specification and the implementation.Radical simplicity means fewer moving parts per unit of capability, not fewer features.DecisionsUse Codex and Claude macOS apps as my primary engineering environmentInputsAgent2Agent (A2A) Protocol documentationAdobe Firefly AI Assistant and agentic creative controlsDemystifying evals for AI agentsClaude Code Review pricingClaude Managed Agents overviewEffective context engineering for AI agentsThe Art Basel and UBS Global Art Market Report 2026Art brutCheap Expertise: Mapping and Challenging Industry Perspectives in the Expert Data Gig EconomyAI Training DomainsUse First Draft with Figma AIFigma MakeHAAI (human-to-agent-to-agents interface) as working termWhat is the future of AI-generated art?Interaction-Augmented Instruction: Modeling the Synergy of Prompts and Interactions in Human-GenAI CollaborationWork That Fits YouGuidelines for Human-AI InteractionMixed Initiative Generative AI Interfaces"Humans welcome to observe": A First Look at the Agent Social Network MoltbookBuy it in ChatGPT: Instant Checkout and the Agentic Commerce ProtocolIntroducing ChatGPT agent: bridging research and actionIntroducing apps in ChatGPT and the new Apps SDKCodex pricingCodex profile and activity surfacesIntroducing deep researchEvaluation best practicesIntroducing workspace agents in ChatGPTTrain the Next Generation of AI as a FreelancerGig Workers Behind AI Face 'Unfair Working Conditions,' Oxford Report FindsObservationsAgent work is becoming supervisory: the scarce act is increasingly deciding what to ask, inspect, accept, or redirect, not typing the implementation.Frontier models such as GPT-5.5 and Claude Opus 4.7 are now good enough at detailed, complex instruction following that a well-structured harness can behave like a practical workflow engine.GitHub has scaling pressure that can show up as a broken experience through intermittent errors, and agent-generated repository volume is one likely contributor.PostsPrompting disciplinesThe all-domain gig economyThe site is the instrumentPredictionsBy 2027-10-01, at least 50% of professional software developers in a credible public developer survey will report that an agent desktop app, agent workbench, or open source equivalent is their primary coding environment rather than a traditional IDE.By 2027-05-14, at least one major AI assistant or cloud platform will launch a public or partner agent marketplace/directory where users can discover, install, rent, or invoke third-party or provider-curated agents, positioning agents as the next distribution layer after apps, plugins, and GPT-style stores.By 2027-02-01, at least one major AI platform will price off-the-shelf or rentable agents using an opaque blended unit such as credits, intelligence units, or agent capacity, where the customer-facing meter combines model choice, token use, tool calls, data access, execution speed, and runtime rather than exposing tokens or wall-clock time as the primary billable unit.By 2030-12-31, Art Brut and adjacent Outsider Art will have materially repriced as an investment target, with public market commentary linking its appeal to the scarcity of uncontrolled, visibly human making in a culture saturated by high-fidelity generative AI imagery.By 2028-06-30, OpenAI will launch or publicly announce a ChatGPT/Codex-centered identity and interaction layer where humans, user-owned agents, and third-party apps or agents have persistent addressable identities and can discover, share, delegate to, or invoke one another inside OpenAI-owned surfaces.By 2027-06-30, a recognizable agent resource layer will exist in which user-facing orchestrator agents act as human-to-agent-to-agents interfaces and delegate work to specialized remote worker agents or harnesses through A2A, an A2A-compatible protocol, or a directly comparable agent-to-agent contract.By 2027-05-14, at least one major AI platform will offer rentable, preconfigured agent harnesses for specific professional task domains, priced by credits, token consumption, execution time, completed work units, or a mixture, so that a user or organization can launch a specialized task agent without building and operating the harness from scratch.ProjectsAgent HarnessCRAN/Eflaming.codesFlamliSvelteKit RuneLayerThinking LabsProvenanceAI provenance for derived claim prompt-is-spec-is-implementationAI provenance for derived claim radical-simplicity-fewer-parts-per-capabilityAI provenance for derived claim agent-runtime-substitutes-for-platformAI provenance for derived claim decision-logic-as-natural-languageAI provenance for derived claim agent-harness-absorbs-complexityAI provenance for derived claim cost-shift-from-infra-to-writing-favorableAI provenance for derived claim agent-systems-collapse-surface-areaThoughtsThe agent harness is the new IDEAgent harnesses are state machinesAgent harnesses will move onto the shelfAgent surfaces are feedback-loop shapedAgents enable radical simplicityAgents lift data journalism up the stackThe harness makes agentic code hold upApp stores will evolve into agent marketsDesign will be seed drivenExpensive AI code review is not insaneOpenAI may want the social layer for humans and agentsIt is starting to feel strange when the work stops as I sleep