# SuperHarness > **Self-containment note (R20):** the external documents this report refers to are vendored under `canon/` as of 2026-07-05. The report's citations record what it read when it was written, and they stay verbatim; to follow one as a live pointer, resolve the doc under `canon/`. :::animation HERO **HERO: the organs of a harness** - **What it shows:** a stylized organism filling the frame, its labeled organs lighting in sequence: prompts at the core, memory and reference as paired lobes, workflows as the skeletal graph, hooks as a reflex layer, loops as the visible heartbeat, operations as the nervous system threading everything - **Narrative role:** sets the thesis; this is the share/card thumbnail - **What it teaches:** a harness is a body, and it lives only when every organ works - **Intended impact:** the reader stops thinking "prompt plus model" and starts thinking anatomy ::: | Field | Value | |---|---| | Project | SuperHarness | | Looikos cluster | Infrastructure & Agent Platforms (the harness layer itself) | | One-line | The brand under which Andy teaches harness engineering, releases stress-tested open-source Hermes-based harness plugins, and sells multi-five-figure harness buildouts. The renamed, recentered successor to Symphony AGI, anchored on the domain name Andy secured. | | Status | Active brand direction as of 2026-07-03. SUPERSEDES `symphony-agi.md` as the brand identity; that deck's market read, valuation comps, personas, and feature-factory analysis remain in force and are referenced here, not repeated. The Hermes/Harness V2 build continues under this name. | | Existing code | The Harness V2 build (`HARNESS_V2_CONSOLIDATED_BRIEF.md`, `HARNESS_V2_HANDOFF.md`, Hermes scaffold, Aion UI); the live harness practice running every Looikos lane; WikiDesignCo as the memory layer under construction (wikidesignco deck §7b) | | Desk | desk-infra | | Coverage | VERIFIED on the seed (Andy's 2026-07-03 directives, first-party), VERIFIED on the loop-engineering market context (2026 practitioner literature, cited), CARRIED-VERIFIED where the Symphony AGI deck's cited reads transfer (tagged inline), INFERRED on tier economics | | Date | 2026-07-03 | --- ## Nine-rung frame (this research task) - **Purpose (the rails):** give the ecosystem the depth to build and run every brand with agents rather than headcount. SuperHarness is the apparatus that purpose runs on, now also the public brand that teaches and sells the apparatus. - **Mission (rung 1):** convert Andy's 2026-07-03 SuperHarness directives plus the standing Symphony AGI research into the deck the rebrand operates from, so the name change lands as a strategic recentering rather than a label swap. - **Objective (rung 2):** a finished deck at `looikos-decks/superharness.md`, supersession explicit, the harness anatomy enumerated as the curriculum spine, loop engineering grounded in the external literature, graded by the Chief. - **Initiative (rung 3):** the 2026-07-03 workspace-manifest doc completion run (the two genuine gaps were data-monastery and superharness; this deck closes the second). - **Project (rung 4):** the looikos-decks corpus, one current-voice deck per ecosystem company. - **Task (rung 5):** this one deck, authored by the Chief per operator directive, deck skeleton and voice floors applied. - **Action (rung 6):** A1 ingest the 2026-07-03 directives and the symphony-agi deck (done). A2 ground loop engineering externally (done, WebSearch; Perplexity MCP down at authoring time). A3 incremental section writing. A4 voice self-check. A5 register in the workspace manifest and memory. - **Decision (rung 7):** what carries over versus what recenters (heuristic: the Symphony AGI market/valuation/persona work is current and cited, so reference it and write only the deltas, per the one-source-many-surfaces rule; authority: Chief, within lane). Deck depth (heuristic: density over length; a rebrand deck that duplicates a live 10k deck is a Disconnection defect; authority: Chief). - **Data (rung 8):** N/A (this doc is the artifact). It later seeds the metagraph as BrandDeck:SuperHarness with a SUPERSEDES edge to BrandDeck:SymphonyAGI, bi-temporally stamped. - **Event (rung 9):** N/A (this doc is the artifact). Deck written to disk, Chief grade recorded. ## 1. What it is (the one-paragraph truth) SuperHarness is the harness engineering brand of the Looikos ecosystem: the public identity under which Andy teaches the discipline, releases stress-tested open-source plugins for the Hermes scaffold, and sells multi-five-figure harness buildouts to organizations whose agentic ambitions have outgrown their duct tape. A harness, in this vocabulary, is everything around the model that turns raw intelligence into dependable work. The brand's central claim is anatomical: the industry is still arguing definitions and semantics, but every serious harness resolves into the same set of organs (prompts, skills, personas, model selection, hooks, reference material, templates, agent SOPs, data models, integrations, memory, evaluation, observability, continuous integration, workflows, loops, error handling, navigation, operations), and the engineering practice that treats them as one body is what separates a demo from an operation. The name is new as of 2026-07-03 and anchored on the domain Andy secured. The substance beneath it is the live Harness V2 build the Symphony AGI deck documented, recentered from "orchestration platform" toward "the discipline, taught and sold." Symphony AGI's conducting philosophy survives intact inside it: you conduct a swarm by understanding what the piece is for, not by adding players. :::animation 1 **ANIMATION 1: demo versus operation** - **What it shows:** two side-by-side rigs built from the same prompt-plus-model core; the left rig runs bare and buckles the moment a second task arrives, the right rig has the harness organs wired around the same core and absorbs the same load without a stumble - **Narrative role:** anchors the central claim of §1, that the harness is what separates a demo from an operation - **What it teaches:** the model is the smallest part; the apparatus around it is what makes work dependable - **Intended impact:** the reader stops crediting the model for reliability and starts crediting the harness ::: ## 2. Andy's seed, expanded (the 2026-07-03 recentering) **Andy's framing (2026-07-03 directives, condensed with fidelity):** SuperHarness is "a Hermes-based harness" brand. The process he designed and is building for himself is powered by OpenAI's Codex as the most powerful coding model, with Claude Code as the orchestrator and personal-assistant layer (driven through the desktop app or through Warp in the terminal), and open source carrying the heavy lifting and parallel execution when multiple perspectives are wanted. A harness breaks down into coherent parts with specialized systems in each: the prompts agents.md soul.md (readmes, "a ton of different markdown files"), skills, personas, model understanding and testing, hooks, reference materials and examples, Jinja templates and the process for managing and updating them, guides and SOPs for agents, data models grounded in Pydantic-as-IR (the discipline ScatterModel productizes), integrations across the agent protocols (MCP, A2A, ACP, ADK, GUIs, "there's so many of them"), memory (with WikiDesignCo as part of it), evaluation rubrics, observability, continuous integration pipelines and feature factories, workflows (orchestrations, LangGraph, DAG-thinking like Airflow), loops ("the modern version of prompts... a big part of what makes agents so powerful now"), errors and retries for reliability and durability, navigation (CLI, grep, APIs, "they need to navigate it somehow"), and operations (an operational ledger with the nine-stage hierarchy, "kind of like a CRM... built in directly to harnesses"). His conviction, in close paraphrase, is that it's so early people are still arguing about semantics, but all of these are going to be vital, "like the parts of a cell, the organs of a body." The brand builds notoriety by intermittently open-sourcing stress-tested pieces, and monetizes as teaching plus "multi five-figure harness buildouts." **Reading between the lines.** Three moves sit inside the recentering. First, the rebrand shifts the product from platform to practice. Symphony AGI was framed as the thing you run on; SuperHarness is framed as the thing you learn and hire. That's a sharper wedge into the same market the Symphony AGI deck mapped, because the buyer's trust problem (a hundred wrapper demos, a fear of science projects) is answered better by a named discipline with a public body of work than by another platform pitch. The platform stays on as the reference implementation the practice teaches from, which is the move the whole ecosystem runs: the working artifact is the demonstration. :::animation 2a **ANIMATION 2a: platform becomes practice** - **What it shows:** a boxed product labeled PLATFORM turns inside out; its walls fold back to reveal a lit workshop where the same components hang on the wall as teachable modules, tools a person picks up rather than a box they rent - **Narrative role:** anchors the first recentering move, platform to practice - **What it teaches:** the rebrand sells the discipline you learn and hire, not the thing you run on - **Intended impact:** the reader sees why a named practice answers the buyer's trust problem better than another platform pitch ::: Second, the three-part Hermes setup assigns model classes to seats instead of betting the brand on any one model. Codex takes the heavy-coder seat because it's the strongest pure implementer; Claude Code takes the orchestrator seat because holding context, dispatching, and grading is its home ground; open-source workers take the parallel fan-out seats because multiple cheap perspectives beat one expensive one on wide problems, with a stronger model synthesizing the winners. The scaffold is the constant and the seats are swappable, which is the same model-agnostic credit-quality argument the Symphony AGI deck made in its finance angle (CARRIED-VERIFIED, symphony-agi.md §3a): the harness proven on the cheaper model upgrades in one line. Third, the anatomy is the curriculum, and the curriculum is the moat. Each organ is a teachable module, a sellable buildout component, and an open-sourceable plugin surface, which means the brand's three revenue motions (teach, release, build) all decompose along the same lines. A competitor can copy any single released plugin; copying the assembled anatomy, with the tested judgments about how the organs compose under production load, requires the operational corpus, and that corpus lives in WikiDesignCo where it compounds privately. :::animation 2b **ANIMATION 2b: the copyable plugin, the uncopyable assembly** - **What it shows:** a rival lifts one glowing plugin cleanly off the anatomy and clicks a perfect copy into their own rig; then they reach for the whole assembled body and their hands pass through it, because the connective tissue of tested compose-judgments lives in a private corpus they cannot reach - **Narrative role:** anchors the third recentering move, the anatomy as the moat - **What it teaches:** single parts are free to copy, the assembled judgment about how they compose is not - **Intended impact:** the reader understands why open-sourcing pieces strengthens rather than leaks the moat ::: :::animation 2 **ANIMATION 2: the Hermes trio** - **What it shows:** three labeled engines docking into one scaffold: CODEX into the heavy-coder bay, CLAUDE CODE into the orchestrator seat, a rack of open-source workers fanning out to the sides; a task visibly splits across the rack, attempts stream back, the orchestrator routes the winner to Codex for the deep build - **Narrative role:** anchors the scaffold-over-model claim - **What it teaches:** the right model in the right seat, and the scaffold outlives any single engine - **Intended impact:** the reader sees model churn as a seat swap, not an existential rewrite ::: ## 3. The anatomy of a harness (the curriculum spine) The list of organs is the brand's load-bearing content: Andy's decomposition, assembled from building these systems daily. Each organ is a real subsystem with its own engineering practice. :::animation 3a **ANIMATION 3a: the organs enumerated** - **What it shows:** the harness organism from the hero returns, and this time each organ is named and lit one at a time in the order the section teaches them (prompts, skills, personas, model selection, hooks, reference, templates, SOPs, data models, integrations, memory, evaluation, observability, CI, workflows, loops, errors, navigation, operations), each glowing then settling into the whole - **Narrative role:** anchors the top of the anatomy section, the curriculum spine - **What it teaches:** the harness is a fixed set of nameable subsystems, each its own engineering practice - **Intended impact:** the reader gets the full map before the detailed walk, so no organ feels arbitrary ::: **Prompts** are the tissue everything grows from: agents.md soul.md readmes, operating contracts, the dozens of markdown files that define who an agent is and how it should think. They're versioned artifacts with owners and review discipline, not strings buried in code. **Skills** package procedural knowledge into loadable units, keeping the base prompt lean while making deep procedure available on demand. **Personas** shape tone and judgment per role, and they make the difference between five agents that all sound like the base model and a crew whose members can be trusted with different work. **Model selection and testing** treats the roster as a portfolio decision, verified by testing rather than vibes and revisited as the frontier moves; Hermes exists because no single model wins every seat. **Hooks** are the mechanical enforcement layer: code that fires on tool use and writes, blocking the em dash, rejecting the invalid model ID, catching the deferral. Prose rules bind an agent that chooses to follow them; hooks bind everyone, and anything load-bearing eventually graduates from instruction to hook. **Reference material** is curated context, and getting the right reference into the right context at the right moment is a retrieval problem, one of several places the harness plugs into WikiDesignCo. **Templates** (Jinja, for everything that repeats) make output shapes deterministic where determinism is cheap, with template lifecycle managed as its own process because a stale template silently propagates a stale standard. **Guides and SOPs for agents** are operating procedures written for machine readers: how to hand off, how to report, what evidence a claim requires, with the nine-rung hierarchy, which runs from Mission to Event, as the master SOP. **Data models** ground every structure in Pydantic as the intermediate representation, so every message and artifact an agent produces has a typed shape that survives translation across storage and transport; ScatterModel is the sister asset that productizes this discipline, one typed model projecting from a single IR to database, backend, frontend, and templates (see the scatter-model deck). **Integrations** answer how the harness connects to the world, and the answer is plural: MCP as today's workhorse, A2A for agent-to-agent traffic, ACP, ADK-style toolkits, plain APIs, CLIs, SDKs, sometimes a GUI driven directly. Knowing which protocol fits which seam is a core harness-engineering judgment. **Memory** is layered by time and truth, with WikiDesignCo as the deep layer: the corpus, the knowledge bases, and the metagraph agents query before acting and write back to after. A harness without a memory architecture repeats its mistakes with great confidence. :::animation 3b **ANIMATION 3b: memory layered by time and truth** - **What it shows:** an agent about to act pauses and reaches down through stacked memory strata (a fast working layer, a reference layer, and the deep WikiDesignCo corpus with its metagraph) to pull the one relevant prior fact up into its context, acts, then writes the outcome back down into the deep layer as a new typed record - **Narrative role:** anchors the memory organ and its WikiDesignCo plug inside the anatomy - **What it teaches:** memory is architecture, layered by recency and by truth, not a single dumped window - **Intended impact:** the reader sees why a harness without a memory layer repeats its mistakes confidently ::: **Evaluation rubrics** define what good means before work begins, and in a loop-driven harness the evaluator is most of the system's real intelligence. **Observability** makes the harness auditable, LogFire spans on every operation of consequence, under the rule that what isn't in the traces didn't happen. **Continuous integration** keeps the harness itself improving: prompts, skills, and hooks flow through the same review discipline as code, because they're code, and the feature factory (the Symphony AGI deck's productized unit, CARRIED-VERIFIED) is this organ at domain scale. **Workflows** structure multi-step work as explicit graphs, LangGraph orchestrations with DAG-thinking a data engineer knows from Airflow; determinism lives in the workflow, judgment lives inside nodes. :::animation 3c **ANIMATION 3c: determinism in the graph, judgment in the node** - **What it shows:** a directed graph of fixed edges lights up as the deterministic skeleton, its arrows firing in a set order; then one node zooms open to reveal a reasoning agent making a genuine call inside, while the graph around it stays rigid and predictable - **Narrative role:** anchors the workflows organ, the split between structure and reasoning - **What it teaches:** the workflow holds the deterministic shape, the node holds the judgment, and mixing them is the failure - **Intended impact:** the reader learns where to put certainty and where to allow reasoning ::: **Errors and retries** are the durability organ: retry policies, idempotency, loud typed failure events, and never a fallback that masquerades as success. A harness earns trust exactly as fast as its failure handling deserves it. **Navigation** is how agents move through their world (a shell with grep, filesystem access, an API surface, a browser), and mismatched navigation is a silent capability killer. **Operations** closes the anatomy: the operational ledger that knows the task, the project, the owner, and the status, structured by the same nine rungs. The ledger is CRM-shaped machinery built for agents as first-class users, the territory PRD Master develops into a product (see the prd-master deck), because a harness without an operational ledger can't be managed, only watched. :::animation 3d **ANIMATION 3d: the operational ledger for agents** - **What it shows:** a live ledger surface where each row is a unit of agent work, its nine rungs from Mission to Event filling column by column as the work proceeds, owner and status updating in place, so the whole fleet's activity reads at a glance the way a CRM reads a sales pipeline - **Narrative role:** closes the anatomy walk on the operations organ - **What it teaches:** agents need a first-class operational record structured by the nine rungs, or the fleet can only be watched, not managed - **Intended impact:** the reader sees operations as the organ that turns a swarm into something governable ::: :::animation 3 **ANIMATION 3: instruction graduates to hook** - **What it shows:** a prose rule written on a card repeatedly failing to stop a rushing stream of agent actions; the card then forges into a mechanical gate bolted across the stream, and the next violating action visibly bounces off it - **Narrative role:** anchors the enforcement claim inside the anatomy - **What it teaches:** anything load-bearing graduates from instruction to hook, because hooks bind everyone - **Intended impact:** the reader feels the difference between advice and architecture ::: ## 4. Loops (the organ the era is named after) Andy weights loops heaviest of all the organs, and the industry has just caught up to why. Loop engineering is the emerging discipline of designing the cycle that drives agents (act, observe, evaluate, decide, repeat) rather than prompting each step by hand, and the 2026 practitioner literature now names it as the successor layer to prompt engineering: the highest-leverage design surface has moved from the words you type to the cycle that keeps an agent productive, verified, and on-goal, including while you sleep (VERIFIED, O'Reilly Radar "Loop Engineering" and the mid-2026 guides, retrieved 2026-07-03). Andy's version of the loop is specific, and it's the one this brand teaches. A fresh agent makes an attempt against pre-established criteria and publishes its work. A reviewer agent evaluates the attempt against those criteria. A third agent reads the review, the original, and the requirements, does its own research, and makes a better attempt. Each participant runs with a lean, purpose-sized context instead of one bloated window, which is why loops produce more quality per token than monolithic sessions: the loop does its action, completes, and keeps going, incrementally and iteratively improving. The external literature lands on the same structural conclusion the practice found first: the verifier is the bottleneck and the crown jewel, because the generator retries cheaply forever, so the quality of the evaluation criteria decides whether all that motion compounds into value (VERIFIED, same sources). Loops are to this era what prompts were to 2023, and the evaluation-rubric organ is where a SuperHarness buildout earns its fee. :::animation 4a **ANIMATION 4a: the verifier is the bottleneck** - **What it shows:** a generator glyph fires attempt after attempt almost for free, a cheap fountain of tries; beside it a single verifier gate is the narrow point everything must pass, and its criteria bar is what actually rises; sharpen the gate and the whole stream's quality climbs, blur it and the fountain of tries compounds into nothing - **Narrative role:** anchors the loops section's second claim, that the evaluator holds the real intelligence - **What it teaches:** the generator retries cheaply forever, so the quality of the evaluation criteria decides whether motion becomes value - **Intended impact:** the reader relocates their effort from generating to specifying what good means ::: :::animation 4 **ANIMATION 4: the loop beats the marathon** - **What it shows:** a circular track with four stations (ATTEMPT, PUBLISH, REVIEW, IMPROVE); a fresh agent glyph spawns with a small clean context meter, completes its arc, despawns; the next glyph spawns standing on the prior work; a fixed criteria bar at center climbs each revolution while every context meter stays small - **Narrative role:** anchors the loops-replace-prompts claim - **What it teaches:** fresh agents on a well-built loop beat one exhausted agent, at lower token cost - **Intended impact:** the reader wants their own loop by the third revolution ::: ## 5. What carries over from Symphony AGI (referenced, not repeated) The Symphony AGI deck (2026-06-21) remains the standing research for everything the rebrand doesn't change, so that research lives in one place. SuperHarness carries over in full, and cites from there, the conducting design philosophy, which derives the topology from the domain's intent, not from imposed volume; the three-angle valuation with its comps (Cognition, Imbue, Replit, Cursor, Zhipu, the MosaicML strategic-multiple logic) and the credit-quality argument for self-hosted, model-agnostic open-core; the artifact-marketplace third door, which treats every reusable unit of agent engineering as a versioned, cataloged asset, a position the incumbents won't occupy; the feature factory as the productized unit; and the five personas with their PST world model. Where that deck says Symphony AGI, read SuperHarness. The metagraph records the supersession as a bi-temporal edge, so the old belief stays queryable as history instead of being erased. The ecosystem treats every renamed belief that way. :::animation 5 **ANIMATION 5: supersession, not erasure** - **What it shows:** a node labeled SYMPHONY AGI does not vanish; a new node SUPERHARNESS lights beside it and a labeled SUPERSEDES edge draws between them, both timestamped, so the old belief stays queryable as history while the new one carries current truth - **Narrative role:** anchors the carry-over section's premise, that the rename is a recentering not a reset - **What it teaches:** the ecosystem records renamed beliefs as bi-temporal edges, keeping the prior research live and cited - **Intended impact:** the reader trusts that nothing researched under the old name was thrown away ::: The rebrand adds four deltas on top of that base. The brand's center of gravity is the practice (teach, release, build) rather than the platform; the anatomy is the public curriculum and the buildout checklist; loops are promoted from an implementation detail to a headline discipline; and the WikiDesignCo dependency is now a build in flight rather than a plan (see the wikidesignco deck), which converts the memory organ from aspiration to integration work. ## 6. Business model and the WikiDesignCo relationship Revenue stacks in three tiers, and the tiers are the same three motions the anatomy decomposes into. The buildout tier sells bespoke harness engineering at multi-five-figure engagements: an organization with real workloads gets a Hermes-based harness designed around the full anatomy, with the evaluation rubrics, the observability, and the operational ledger that make it maintainable after the engagement ends, priced against the fear it resolves (the science project that never ships). The teaching tier packages the discipline itself, the anatomy as modules and loop engineering as the flagship, with Data Monastery as the neighboring teaching brand for the data layer beneath (see the data-monastery deck). The open-source tier is the top of the funnel rather than a revenue line: internally stress-tested Hermes plugins released deliberately and intermittently, each one a proof artifact aimed at the engineers who influence buildout purchases. The tier economics are INFERRED pending real pipeline; the motion structure is VERIFIED as Andy's stated design. :::animation 6a **ANIMATION 6a: three motions, one anatomy** - **What it shows:** the harness anatomy sits at the center and three revenue arrows draw off the same organs: an open-source funnel releasing single lit plugins at the top, a teaching tier lifting organs out as curriculum modules, and a buildout tier assembling the whole body for one client, all decomposing along the identical organ lines - **Narrative role:** anchors the revenue-model claim that teach, release, and build share one spine - **What it teaches:** the three revenue motions are the same anatomy read three ways, not three separate businesses - **Intended impact:** the reader sees the open-source release as the top of a funnel, not a giveaway ::: The WikiDesignCo relationship runs in both directions and is the deepest dependency in the ecosystem. Downward, every organ that touches knowledge runs on the platform: memory queries the corpus and the metagraph, reference material is retrieved from knowledge bases through their MCP connections, templates and rubrics live as versioned assets, and the operational ledger's records land as typed, provenance-bearing data. Upward, the platform is operated by harnesses, so every SuperHarness advance upgrades the crew that runs the corpus. Both share the staging the 2026-07-03 directives set out: the platform is the environment, humans and CLI agents are its users through the same surfaces, Gemini and Vertex carry the in-platform processing for now (LangGraph workflows or direct calls, Gemini being the model that reads video), and resident Hermes-class agents arrive later, low-level processing first, while real-time acting stays with humans and CLI agents. The ContentFactory live instance is the laboratory where that staging earns its keep on a paying customer today. :::animation 6 **ANIMATION 6: the two-way dependency** - **What it shows:** a harness organism and a corpus field connected by two thick flows: knowledge streaming up from the corpus into the harness's memory and reference lobes, and refined work streaming down from the harness's hands into the corpus as new typed assets - **Narrative role:** anchors the SuperHarness-WikiDesignCo relationship - **What it teaches:** the harness runs on the platform and the platform is run by harnesses; each advance upgrades the other - **Intended impact:** the reader understands why these two assets are built as one motion ::: ## 7. Priority read (feeds the value rubric) SuperHarness inherits Symphony AGI's position on the dependency graph as the apparatus every brand is conducted by. It's foundational alongside WikiDesignCo, and the two form the substrate pair: the harness runs the work, and the platform remembers it. The work now is to continue the Hermes/Harness V2 build under the new name, and let the anatomy structure both the internal practice and the first public teaching material. Next, once WikiDesignCo's walking skeleton lands, the work is to wire the memory organ to real workspace knowledge bases via MCP, and cut the first open-source plugin release from a stress-tested internal piece. The thing to watch is the loop-engineering conversation, which is moving fast in public, so the teaching material's framing gets re-grounded against the literature at each authoring pass rather than assumed stable. What to leave alone is unchanged from the Symphony AGI read: never rebuild what the scaffold rents (model runtimes, commodity orchestration primitives). :::animation 7 **ANIMATION 7: the substrate pair** - **What it shows:** two foundation stones set side by side under the whole ecosystem, one labeled SUPERHARNESS (the crew that runs the work) and one WIKIDESIGNCO (the corpus that remembers it); the brand nodes above rest their weight on both, and the Now/Next/Watch/Leave lanes sort into place around them - **Narrative role:** anchors the priority read, SuperHarness as foundational substrate paired with the platform - **What it teaches:** the harness and the platform are the two-stone floor every other brand stands on - **Intended impact:** the reader ranks this brand as substrate, never above the thing it depends on ::: ## 8. The brand's own nine-rung position - **Purpose (rails):** own the discipline of making agents dependable, and make that discipline the ecosystem's public proof of competence. - **Mission (1):** establish harness engineering as a named practice with SuperHarness as its reference voice, taught in public and sold as buildouts. - **Objective (2):** the Hermes build proven across the constellation's real workloads under the SuperHarness name, the anatomy published as the curriculum spine, and the first buildout engagement closed on the strength of released artifacts. - **Initiative (3):** the rebrand-and-recenter from Symphony AGI (2026-07-03), executed as supersession with the research base carried forward. - **Project (4):** the Harness V2 build continuing under this name; definition of done per its consolidated brief. - **Task (5):** a unit a single agent executes, for example authoring one anatomy module or stress-testing one plugin against a constellation workload. - **Action (6):** an atomic operation, for example running one loop revolution against fixed criteria, or releasing one versioned plugin with its documentation. - **Decision (7):** which organ a given failure implicates (heuristic: the anatomy as diagnostic checklist; authority: within lane); when a plugin is stress-tested enough to release (heuristic: survived N real workloads with clean traces; authority: operator). - **Data (8):** the ECS records the practice produces: HarnessConfig, PluginRelease, BuildoutEngagement, LoopCriteria, each a typed Pydantic-IR entity landing in WikiDesignCo. - **Event (9):** the real occurrences captured: a loop revolution completed, a plugin released, a buildout shipped, an evaluation verdict recorded, each with its LogFire span. :::animation 8 **ANIMATION 8: every act lands as a typed record** - **What it shows:** a loop revolution completes and drops a typed token into a ledger (LoopCriteria met), a plugin release drops another (PluginRelease), a buildout closes and drops a third (BuildoutEngagement), each token carrying its LogFire span as a thread of light trailing back to the moment it happened - **Narrative role:** anchors the brand's own bottom rungs, Data and Event, as lived records - **What it teaches:** the practice produces typed, provenance-bearing records for every act, not just outputs - **Intended impact:** the reader sees the discipline as auditable by construction, down to the single event ::: ## 9. Sources Evidence-tag legend: VERIFIED (first-party directive, repo primary, or cited external), CARRIED-VERIFIED (verified in the Symphony AGI deck and carried by reference), INFERRED (reasoned, not confirmed), OPEN (acknowledged gap). - Andy's 2026-07-03 directives (first-party, the seed of this deck: the SuperHarness name and domain, the Hermes trio, the full anatomy enumeration, the loops framing, the teach/release/build model, the platform-as-environment staging). VERIFIED. - `symphony-agi.md` (this corpus): the superseded brand deck; conducting philosophy, valuation comps, artifact-marketplace third door, feature factories, personas. CARRIED-VERIFIED; where it says Symphony AGI, read SuperHarness. - `wikidesignco.md` §7b (this corpus) and `wikidesignco/WDC_PLATFORM_V1_BRIEF.md`: the platform build the memory organ integrates with. VERIFIED. - Loop engineering external grounding (WebSearch 2026-07-03; Perplexity MCP down at authoring time): O'Reilly Radar "Loop Engineering"; MindStudio "What Is Loop Engineering? The New Meta for AI Coding Agents"; tosea.ai "Loop Engineering: A Complete Guide from Prompt to Harness Engineering (2026)"; knightli.com "Loops Replace Prompts." Key transferred findings: the act-observe-verify-iterate structure, the migration prompt-to-context-to-environment-to-loop, and the verifier-as-bottleneck conclusion. VERIFIED as literature. - `wikidesignco/RAW_knowledgebase/19-superharness.md`: the Chief's same-day long-form brief in the older series format; the anatomy prose here is its port into deck grounding. VERIFIED, subordinate to this deck. - Tier economics for teach/release/build: INFERRED pending real pipeline. Coverage statement: VERIFIED-heavy on seed and anatomy, CARRIED on market/valuation/personas, INFERRED on pricing, OPEN on the first external buildout reference case. **Created:** 2026-07-03. Authored by the Chief agent from Andy's directives of the same date. Supersedes `symphony-agi.md` as the brand identity; that deck remains the standing market research.