# Workflow guide

This Area → Workflow → Stage → Role taxonomy applies to native harness
subscriptions (Claude CLI, Codex, Grok CLI, agy/Antigravity) and to API-key Pi
providers (OpenRouter, Nous Portal). Vendor-prefixed model ids in the candidate
lists are dated research candidates, not admission.

## Table of Contents

1. [Software Engineering](#software-engineering)
2. [Design & Experience](#design--experience)
3. [Media Production](#media-production)
4. [Data & Analytics](#data--analytics)
5. [Research & Strategy](#research--strategy)
6. [Business Operations](#business-operations)
7. [Security & Reliability](#security--reliability)
8. [Selection Policy](#selection-policy)
9. [Slug Registry](#slug-registry)
10. [Historical Navigation](#historical-navigation)

---

## Provenance

The workflow and role candidate lists in this guide are inherited research from the snapshot committed at `67e3f99313dadbbd8ef06884be0588fc92970caf` on 2026-09-06. That is a document snapshot date, not a catalog-verification date; current price and capability freshness is unverified. The lists are dated candidates that require live verification before dispatch. They are not certified prices or capabilities, and they are not an eligibility grant.

Existing claims inside preserved candidate lines are research claims, not dispatch policy or proven quality guarantees. Area grouping is navigation and never pools unrelated role quality into one global model ranking. Model, harness, platform, modality, required tools, and personal or work context are routing attributes, not area trees.

---

## Selection Policy

Candidate selection is measured per role. Filter stages are optional and are not mandatory Tier-0 gating. Prefer a provider-native authenticated subscription when Tracking says that harness is eligible. For Gemini candidates (`google/*`), the admitted native subscription harness on this runner is `agy` (Antigravity CLI, `agy -p`), not the OpenRouter provider id, when Tracking lists agy as eligible. Do not change the candidate ids themselves (they are dated research). Evidence and review remain workflow-specific. Both the Fable architecture critic and the Sol CLI critic remain required for this developer assignment runner.

### Cost band reference (dated candidates)

| Tier | Economic Band | Context Window | Primary Models | Core Workload Profile |
|---|---|---|---|---|
| **Tier 0: Filter & Ingestion** | $0.03 - $0.20 / M | 1.0M - 1.31M | DeepSeek-V4-Flash, Qwen-3.7-Flash, GLM-5.3-Flash, GPT-5.6-Luna | High-throughput telemetry, raw log ingestion, triage, video pre-filtering, and initial inbox classification. |
| **Tier 1: Workhorse & Engine** | $0.30 - $1.00 / M | 262k - 1.05M | Gemini-3.8-Flash, Qwen-3-Coder-Plus, DeepSeek-V4-Pro, Devstral-2512 | Routine code generation, multimodal visual QA, Text-to-SQL, AST diff patching, and fast tool dispatching. |
| **Tier 2: Precision & Critic** | $1.25 - $4.00 / M | 200k - 1.05M | GPT-5.6-Sol, Grok-4.6, Claude Sonnet 5, OpenAI o3, DeepSeek-R1, Codex 5.3 | Primary code writing, formal contract generation, concurrency race diagnosis, and causal inference. |
| **Tier 3: Sovereign Architecture** | $5.00 - $15.00 / M | 1.0M - 1.05M | Claude Fable 5.1, Claude Opus 5, GPT-6-Astra-Pro (Escalation only) | Designated architecture critic, executive brief, high-stakes legal redlining, and sovereign RFC review. |

### Developer runner loop

The developer assignment runner executes a gated loop with rework across every software engineering workflow. This is not a seventh stage or an informal convention; it is the enforced loop for every software engineering workflow on this runner:

```text
plan (claude/fable)
  │
  ▼
implement (grok-native)
  │
  ▼
witness (fixed gate: npm run verify)
  │
  ▼
dual critics (review-arch: fable + review-cli: sol)
  │
  ├─ If either critic BLOCK ──────────────┐
  │                                        ▼
  │                                repair (rework_of binding;
  │                                        failover to next eligible
  │                                        authenticated writer)
  │                                        │
  │                                        └─ re-witness + fresh dual review
  │
  ▼ (both critics PASS on exact tree)
accept (just accept binds commit + both PASS records)
  │
  ▼
pull request (five CI jobs: 2 Linux validate, classify, docs, plugin)
```

**Runner loop invariants:**

- **Roles and rotation:** The runner strictly maps assignments to authenticated roster roles (`AGENTS.md` and `.kxm/roster.json`):
  - **Implement / write code (`writer`):** **Grok** (`grok --model grok-4.6`, headless). Currently admitted native writer on this runner. Failover follows attempts-and-relief: after an empty or failed attempt, immediately fail over to the next eligible authenticated writer in the roster (**Qwen** `openrouter/qwen/qwen3-coder-plus` via Pi).
  - **Plan (`planner`):** **Claude Fable** (`claude --model fable`), read-only architecture and permissions planning.
  - **Review architecture (`reviewer-arch`):** **Claude Fable** (`claude --model fable`), designated architecture and permissions critic.
  - **Review CLI / contracts (`reviewer-cli`):** **Codex Sol** (`codex` `gpt-5.6-sol`), designated CLI, protocol, and documentation critic.
- **Fixed witness gate:** `npm run verify` (incorporating tests, typecheck, docs lint, version checks, and generated `dist` match) runs deterministically as the witness gate before critics review. Deterministic gates beat a third model; models propose, gates hold the line. Model critics review contracts, architecture, and documentation; they never replace the deterministic compiler or test runner.
- **Dual-critic quorum:** PR acceptance strictly requires **both** designated critics (**Fable** for architecture/permissions and **Sol** for CLI/contracts) to record a `PASS`. A single critic is at most preliminary triage; neither critic may share a provider with each other or with the writer.
- **Repair back-edge and rework:** If either critic records a `BLOCK`, the runner emits a `repair` assignment bound to the prior attempt via `rework_of`. Every repair round re-runs the fixed witness gate and requires **fresh** dual critic reviews on the resulting tree.
- **Attempts and relief:** Never stop solely because attempts are exhausted or failed. Immediately try the next suggested eligible authenticated model and transfer findings while preserving every attempt, candidate, failed check, and cost record (`AGENTS.md`).
- **Acceptance and auto-merge:** `just accept` binds the exact candidate commit hash and independent Fable + Sol PASS records. Changes are pushed to a feature branch and merged via pull request with auto-merge after all five CI jobs pass.

---

## Slug Registry

Area and workflow slugs are lower-case kebab-case ASCII and do not use numerals as canonical identity. They are documentation identity, declared here separately from display names and from GitHub auto-anchors. Role slugs are reusable specialty slugs: kebab-case, no area or workflow prefix, and no numbers. The same specialty may recur across workflows; a role slug must be unique within a workflow. When a role must be disambiguated, the composite reference is the explicit namespace `area-slug/workflow-slug/role-slug` (for example `security-reliability/investigate-incident/forensic-causal-analyst`). Never use a bare number such as 3.1.2 as identity.

These slugs imply no runtime config, role admission, schema field, CLI behavior, or alias. Role slugs are declared under each role heading.

### Areas

| Area | Slug |
|---|---|
| Software Engineering | `software-engineering` |
| Design & Experience | `design-experience` |
| Media Production | `media-production` |
| Data & Analytics | `data-analytics` |
| Research & Strategy | `research-strategy` |
| Business Operations | `business-operations` |
| Security & Reliability | `security-reliability` |

### Workflows

| Area | Workflow | Slug |
|---|---|---|
| Software Engineering | Build Feature | `build-feature` |
| Software Engineering | Refactor and Repair Regressions | `refactor-repair-regressions` |
| Software Engineering | Stabilize Flaky Tests | `stabilize-flaky-tests` |
| Software Engineering | Design Software System | `design-software-system` |
| Software Engineering | Maintain Documentation | `maintain-documentation` |
| Design & Experience | Build Design System | `build-design-system` |
| Design & Experience | Engineer Mobile Interactions | `engineer-mobile-interactions` |
| Design & Experience | Engineer Terminal Interfaces | `engineer-terminal-interfaces` |
| Design & Experience | Audit Visual Quality and Accessibility | `audit-visual-accessibility` |
| Media Production | Produce Generative Video | `produce-generative-video` |
| Media Production | Repurpose Long-Form Video | `repurpose-long-form-video` |
| Media Production | Edit and Render Media | `edit-render-media` |
| Data & Analytics | Query Business Intelligence | `query-business-intelligence` |
| Data & Analytics | Analyze Dataset | `analyze-dataset` |
| Data & Analytics | Extract and Audit Documents | `extract-audit-documents` |
| Research & Strategy | Prepare Decision Brief | `prepare-decision-brief` |
| Research & Strategy | Structure Negotiations | `structure-negotiations` |
| Business Operations | Automate Tasks | `automate-tasks` |
| Business Operations | Handle Customer Escalations | `handle-customer-escalations` |
| Business Operations | Review Contracts | `review-contracts` |
| Security & Reliability | Investigate Incident | `investigate-incident` |
| Security & Reliability | Patch Vulnerability | `patch-vulnerability` |

---

## Software Engineering

*Slug:* `software-engineering`

**Overview:**

1. **Build Feature:** PRD Ingestion $\to$ System Planning $\to$ API Spec/Contract Authoring $\to$ Workspace Scaffolding $\to$ Core Backend & Data Logic $\to$ Full-Stack UI Implementation $\to$ Tool/SDK Integrations $\to$ Automated Verification.
2. **Refactor and Repair Regressions:** Codebase Smell Analysis $\to$ Modular Decomposition $\to$ Surgical Multi-File AST Transforms $\to$ Static Typing & Contract Verification.
3. **Stabilize Flaky Tests:** Flakiness Root-Cause Extraction $\to$ Deterministic Mock/Async Hardening.
4. **Design Software System:** NFR/SLA Ingestion $\to$ System Topology & Trade-Offs $\to$ Technical RFC Authoring $\to$ STRIDE Threat Modeling $\to$ Storage/Sharding Design $\to$ IaC Cloud Topology $\to$ Chaos/DR Review $\to$ Independent Critic Quorum.
5. **Maintain Documentation:** Accuracy Audit $\to$ Documentation Write $\to$ Witness Gate $\to$ Dual Review Quorum.

---

### Workflow: Build Feature

*Slug:* `build-feature`

**Stages:** System planning → Spec and contract → Scaffold → Backend and data → Frontend → SDK integration

#### Role: Lead Systems Planner

*Slug:* `lead-systems-planner` | *Stage:* System planning

*Domain:* Decomposes requirements into an executable, dependency-ordered DAG of modules, tasks, and verification gates.

1. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Sovereign systems planner; strict boundary isolation and non-overlapping DAG synthesis.
2. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Exceptional price-to-reasoning ratio; detects race conditions and circular dependencies.
3. **`openai/gpt-6-astra-pro`** (1.05M ctx | $10.00 / $50.00) — Enterprise task decomposition and critical-path scheduling across massive PRDs.
4. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Pragmatic engineering focus; rapid modular task decomposition.
5. **`qwen/qwen3.8-max-0902`** (1M ctx | $2.00 / $6.00) — Broad framework knowledge; maps migration orders and build prerequisites.

#### Role: Spec Contract Generator

*Slug:* `spec-contract-generator` | *Stage:* Spec and contract

*Domain:* Authors zero-ambiguity API contracts, type definitions, and schema validation rules before implementation begins.

1. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — Premier schema contract generator; 100% compliant OpenAPI 3.1, Protobuf v3, and JSON schemas.
2. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — High-throughput Zod/Pydantic validation pipeline synthesis.
3. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Fast translation of data dictionaries into TypeScript interfaces.
4. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — 1M-context contract verification across monorepo boundaries.
5. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Reserved for mission-critical authentication and security contract definitions.

#### Role: Scaffold Build Specialist

*Slug:* `scaffold-build-specialist` | *Stage:* Scaffold

*Domain:* Provisions monorepo workspace topologies (Turborepo, Cargo, pnpm), linters, Docker multi-stage builds, and CI pipelines.

1. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Designated native repo builder; resolves path aliases, linkings, and workspace configs cleanly.
2. **`mistralai/devstral-2512`** (262k ctx | $0.40 / $2.00) — Developer-focused open model; produces clean Makefiles, Taskfiles, and tool configs.
3. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — Coordinates cross-package build configs and dependency lockfiles in 1M context.
4. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Complex CI/CD workflows and multi-stage container build optimizations.
5. **`kwaipilot/kat-coder-pro-v2.5`** (262k ctx | $0.74 / $2.96) — Resilient project scaffolding and bundler configuration.

#### Role: Backend Data Implementer

*Slug:* `backend-data-implementer` | *Stage:* Backend and data

*Domain:* Implements domain entities, SQL queries/migrations, ORM persistence, concurrency primitives, and transactional boundaries.

1. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Admitted native writer; passes deterministic verify gates (`npm run verify`) on the first pass with minimal rework.
2. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — 1M-context open-weight anchor; generates performant database layers and CRUD services.
3. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Exceptional database logic, indexing, and edge-case domain implementation.
4. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Algorithmic precision in Rust, Go, Python, and TypeScript concurrent systems.
5. **`kwaipilot/kat-coder-pro-v2.5`** (262k ctx | $0.74 / $2.96) — Multi-file backend logic implementation adhering to project coding patterns.

#### Role: Frontend Fullstack Implementer

*Slug:* `frontend-fullstack-implementer` | *Stage:* Frontend

*Domain:* Constructs stateful UI components, routing, client cache management, responsive Tailwind styles, and accessible interactions.

1. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Industry leader for UI engineering; elegant Tailwind/React component hierarchies and hooks.
2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Fast streaming UI prototyping, token parsing, and JSX/TSX tree generation.
3. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Fast full-stack wiring, shadcn/ui integration, and client-server action implementations.
4. **`bytedance-seed/seed-2.0-code`** (262k ctx | $0.50 / $3.00) — Specialized web engineering model for interactive client components.
5. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Complex client cache mutations (TanStack Query) and optimistic UI state machines.

#### Role: SDK Integration Specialist

*Slug:* `sdk-integration-specialist` | *Stage:* SDK integration

*Domain:* Wires external APIs, authentication flows (OAuth2/OIDC), webhook validation, cloud SDKs, and async worker queues.

1. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — Flawless SDK synthesis, webhook signature validation, and API adapter error handling.
2. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Robust glue code, rate-limiting backoff algorithms, and token refresh loops.
3. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — Ingests full third-party API documentation to emit typed wrapper clients.
4. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Low-cost mapping of disparate vendor payloads into internal domain models.
5. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — High-speed webhook listener and protocol transformation adapter.

---

### Workflow: Refactor and Repair Regressions

*Slug:* `refactor-repair-regressions`

**Stages:** Refactor architecture → Code transform → Equivalence verification

#### Role: Refactoring Architect

*Slug:* `refactoring-architect` | *Stage:* Refactor architecture

*Domain:* Analyzes code coupling, designs clean modular boundaries, and authors step-by-step non-breaking migration plans.

1. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Strict boundary isolation and dependency inversion planning across large codebases.
2. **`z-ai/glm-5.3`** (1.31M ctx | $1.40 / $4.40) — 1.31M context full-repo symbol dependency analysis.
3. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Deep architectural judgment; anticipates runtime blast radius during refactors.
4. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Cost-efficient full-codebase impact analysis.
5. **`moonshotai/kimi-k3`** (1.05M ctx | $3.00 / $15.00) — Multi-package refactoring plan generation.

#### Role: Code Transform Implementer

*Slug:* `code-transform-implementer` | *Stage:* Code transform

*Domain:* Executes surgical multi-file edits, applying design patterns while preserving comments and formatting.

1. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Designated native writer; fast execution of multi-file semantic code changes.
2. **`relace/relace-apply-3`** (256k ctx | $0.85 / $1.25) — Specialized model for deterministic AST code patch merging without syntax corruption.
3. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — High-precision refactoring of complex pointer/generic logic.
4. **`morph/morph-v3-large`** (262k ctx | $0.90 / $1.90) — 4,500 tok/sec high-accuracy mechanical patch application.
5. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Broad contextual refactoring across multi-file boundaries.

#### Role: Semantic Equivalence Verifier

*Slug:* `semantic-equivalence-verifier` | *Stage:* Equivalence verification

*Domain:* Critic-with-gate. Deterministic checks (`tsc --noEmit`, `npm test`, `npm run check`, generated `dist` match) serve as the fixed witness gate and are never delegated to an LLM; the model role reviews public contract compliance, CLI behavior, and behavioral equivalence without replacing the gate ("Deterministic gates beat a third model… models propose; gates hold the line").

1. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — Designated CLI/spec critic; reviews compiler cascades and type-checker cascades against API contracts.
2. **`openai/o3-mini-high`** (200k ctx | $1.10 / $4.40) — Deep symbolic verification that AST refactors maintain behavioral equivalence.
3. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Type-cascade and semantic regression review across package boundaries.
4. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Verifies backwards compatibility and deprecation notices across public APIs.
5. **`mistralai/devstral-2512`** (262k ctx | $0.40 / $2.00) — Fast lint rule and syntax compliance review.

---

### Workflow: Stabilize Flaky Tests

*Slug:* `stabilize-flaky-tests`

**Stages:** Flakiness diagnosis

#### Role: Concurrency Flakiness Detective

*Slug:* `concurrency-flakiness-detective` | *Stage:* Flakiness diagnosis

*Domain:* Untangles timing hazards, async wait conditions, shared state pollution, and unhandled promise rejections.

1. **`openai/o3`** (200k ctx | $2.00 / $8.00) — Unrivaled reasoning on async timing races, microtask queues, and thread contention.
2. **`deepseek/deepseek-r1-0528`** (164k ctx | $0.50 / $2.15) — Exposes hidden execution dependencies and global singleton leaks between tests.
3. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Ingests hundreds of CI failure logs to map statistical failure distributions.
4. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Identifies test runner interactions with underlying event loops.
5. **`openai/gpt-6-astra`** (1.05M ctx | $10.00 / $50.00) — Complex distributed deadlocks and environmental flakiness.

---

### Workflow: Design Software System

*Slug:* `design-software-system`

**Stages:** Distributed architecture → Database topology → Threat modeling → IaC topology → Reliability review → Protocol audit

#### Role: Principal Distributed Architect

*Slug:* `principal-distributed-architect` | *Stage:* Distributed architecture

*Domain:* Evaluates CAP/PACELC trade-offs, consensus protocols (Raft, Sagas), partition boundaries, and authors primary RFCs.

1. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Designated architecture planner; strict fault domain and consensus design.
2. **`openai/o3`** (200k ctx | $2.00 / $8.00) — Formal proof of consensus safety, liveness, and split-brain recovery logic.
3. **`openai/gpt-6-astra-pro`** (1.05M ctx | $10.00 / $50.00) — Multi-tier cloud topology and global multi-region architecture design.
4. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Publication-grade RFC authoring detailing operational trade-offs and team boundaries.
5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Cost-effective evaluation of microservice communication topologies.

#### Role: Scalability Database Topologist

*Slug:* `scalability-database-topologist` | *Stage:* Database topology

*Domain:* Designs relational/NoSQL schemas, horizontal sharding keys, caching topologies (Redis/Valkey), and zero-downtime migration plans.

1. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Primary workhorse for high-scale sharding and replication modeling.
2. **`openai/gpt-6-astra-pro`** (1.05M ctx | $10.00 / $50.00) — Distributed multi-master database topology and write-conflict resolution.
3. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Strict consistency vs eventual consistency boundary auditing.
4. **`qwen/qwen3.8-2.4t-a95b`** (1.05M ctx | $2.00 / $6.00) — Partitioning, indexing, and high-throughput query topology.
5. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Pragmatic cache tiering (Redis/Dragonfly) and read-replica strategies.

#### Role: Security Architect Threat Modeler

*Slug:* `security-architect-threat-modeler` | *Stage:* Threat modeling

*Domain:* Conducts STRIDE threat assessments, reviews IAM least-privilege policies, mTLS boundaries, and zero-trust architectures.

1. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Designated architecture/permissions critic; zero-trust network modeling.
2. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — Designated CLI/spec critic; auth token and protocol threat analysis.
3. **`openai/gpt-6-astra-pro`** (1.05M ctx | $10.00 / $50.00) — Enterprise IAM and cloud privilege escalation modeling.
4. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — STRIDE threat matrix synthesis and compliance verification.
5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Full-infrastructure vulnerability surface analysis.

#### Role: IaC Cloud Topology Engineer

*Slug:* `iac-cloud-topology-engineer` | *Stage:* IaC topology

*Domain:* Generates declarative Terraform/OpenTofu, Pulumi, Kubernetes CRDs, and VPC networking definitions.

1. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Fast, deterministic Terraform/Kubernetes manifest implementation.
2. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — 1M context whole-infrastructure state dependency resolution.
3. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Complex cloud provider networking and peering configurations.
4. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — High-speed validation of cloud resources and linting.
5. **`mistralai/devstral-2512`** (262k ctx | $0.40 / $2.00) — Lightweight Helm chart and Docker Compose generator.

#### Role: Reliability Chaos Reviewer

*Slug:* `reliability-chaos-reviewer` | *Stage:* Reliability review

*Domain:* Evaluates single points of failure, RPO/RTO metrics, cascading timeouts, circuit breakers, and active-active failovers.

1. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Principal disaster recovery and cascading failure reviewer.
2. **`openai/o3`** (200k ctx | $2.00 / $8.00) — Probabilistic modeling of MTTR, MTBF, and network partition recovery.
3. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Full-topology chaos injection scenario planning.
4. **`openai/gpt-6-astra-pro`** (1.05M ctx | $10.00 / $50.00) — Multi-region disaster recovery runbook formulation.
5. **`moonshotai/kimi-k3`** (1.05M ctx | $3.00 / $15.00) — Cross-datacenter failover verification.

#### Role: Architecture Critic Protocol Auditor

*Slug:* `architecture-critic-protocol-auditor` | *Stage:* Protocol audit

*Domain:* Independent peer-review; challenges unvalidated assumptions, prevents architectural drift, and verifies contracts.

1. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — **Designated Architecture & Permissions Critic** (Mandatory for PR acceptance).
2. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — **Designated CLI, Protocol & Contracts Critic** (Mandatory for PR acceptance).
3. **`z-ai/glm-5.3`** (1.31M ctx | $1.40 / $4.40) — 1.31M context full-stack protocol audit.
4. **`openai/gpt-6-astra-pro`** (1.05M ctx | $10.00 / $50.00) — Enterprise RFC compliance verification.
5. **`qwen/qwen3.8-max-0902`** (1M ctx | $2.00 / $6.00) — Independent open-weights architectural validation.

---

### Workflow: Maintain Documentation

*Slug:* `maintain-documentation`

**Stages:** Accuracy audit → Documentation write → Review

#### Role: Documentation Accuracy Auditor

*Slug:* `documentation-accuracy-auditor` | *Stage:* Accuracy audit

*Domain:* Read-only planner; audits developer and operator documentation against CLI `--help` outputs, source environment variables, tool schemas, and repository configuration before documentation authoring begins.

1. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Designated architecture and permissions planner; audits authority, permissions, and lifecycle models against reality.
2. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — High-throughput cross-referencing between source code symbols and markdown documentation.
3. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — Precision auditing of CLI commands, flags, schema definitions, and environment variables.
4. **`qwen/qwen3.8-max-0902`** (1M ctx | $2.00 / $6.00) — Broad codebase auditing against guides, tutorials, and operational manuals.
5. **`z-ai/glm-5.3`** (1.31M ctx | $1.40 / $4.40) — Long-context consistency verification across multi-document doc trees.

#### Role: Documentation Writer

*Slug:* `documentation-writer` | *Stage:* Documentation write

*Domain:* Authors concise, accurate developer and operator documentation, ensuring clear terminology, correct command flags, and consistent formatting without marketing drift.

1. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Admitted native writer; rapid, structured markdown authoring adhering to strict repository conventions.
2. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — High-context technical writer; coordinates cross-file references across large documentation sets.
3. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Fluid technical documentation authoring with clear instructional hierarchy.
4. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Accurate technical guides, CLI flag references, and runnable examples.
5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Cost-effective multi-page documentation generation and reference expansion.

#### Role: Documentation Reviewer

*Slug:* `documentation-reviewer` | *Stage:* Review

*Domain:* Critic-with-gate. Deterministic gates (`npm run lint:docs`, `test/docs-copy.test.ts` docs brake, and `npm run check`) serve as the fixed witness gate and are never replaced by LLMs; critics conduct independent peer review where the architecture critic verifies authority and permission wording, and the CLI critic verifies command/flag accuracy and documentation linting.

1. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — **Designated Architecture & Permissions Critic** (Mandatory for PR acceptance); verifies security, governance, and authority wording.
2. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — **Designated CLI, Protocol & Contracts Critic** (Mandatory for PR acceptance); verifies CLI command accuracy, flag syntax, and documentation formatting.
3. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Rigorous editorial review for clarity, technical completeness, and tone consistency.
4. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Systematic verification of cross-document links and anchor consistency.
5. **`mistralai/devstral-2512`** (262k ctx | $0.40 / $2.00) — Rapid lint rule compliance and syntax verification.

---

## Design & Experience

*Slug:* `design-experience`

**Overview:**

1. **Build Design System:** W3C DTCG Token Modeling $\to$ Style Dictionary v4 AST Engine $\to$ Headless Component Primitives $\to$ Compound Variants (CVA).
2. **Engineer Mobile Interactions:** Harmonic Spring Physics $\to$ Reanimated v3 UI-Thread Worklets $\to$ SwiftUI/Compose Gestures $\to$ CoreHaptics Waveforms $\to$ 120Hz LTPO Profiling.
3. **Engineer Terminal Interfaces:** Raw Mode Protocol $\to$ TrueColor Double-Buffering $\to$ The Elm Architecture (Bubbletea) / Ratatui Layouts $\to$ Terminal Signal Trapping.
4. **Audit Visual Quality and Accessibility:** Snapshot Ingestion $\to$ Perceptual Diffing (SSIM) $\to$ WCAG 2.2 / APCA Contrast Math $\to$ Concentric Radius / Optical Polish.

---

### Workflow: Build Design System

*Slug:* `build-design-system`

**Stages:** Token architecture → Component synthesis

#### Role: Design Token Theme Architect

*Slug:* `design-token-theme-architect` | *Stage:* Token architecture

*Domain:* W3C DTCG token schema authoring, Style Dictionary v4 AST pipelines, OKLCH/Display-P3 scales, and multi-platform packaging (Web, Swift, Compose).

1. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Master design token taxonomist; color space (OKLCH) and scale hierarchy.
2. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Multi-platform token transformation (W3C Design Tokens to CSS/iOS/Android).
3. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — Monorepo theme implementation and code generation.
4. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Fast token validation and contrast ratio auditing.
5. **`openai/gpt-5.4`** (1.05M ctx | $2.50 / $15.00) — Enterprise design system cross-platform consistency.

#### Role: Web Component Synthesizer

*Slug:* `web-component-synthesizer` | *Stage:* Component synthesis

*Domain:* Modern React 19, Tailwind CSS v4 `@theme`, Radix UI headless primitives, polymorphic slot patterns, and CVA variant systems.

1. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Clean React 19, Tailwind, and accessible Radix/Aria primitives.
2. **`bytedance-seed/seed-2.0-code`** (262k ctx | $0.50 / $3.00) — Highly efficient component code synthesizer.
3. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Fast state machine implementation and micro-interaction code.
4. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Complex layout virtualization and DOM performance tuning.
5. **`deepseek/deepseek-v4-pro`** (1.05M ctx | $0.63 / $1.25) — Bulk UI component library scaffolding.

---

### Workflow: Engineer Mobile Interactions

*Slug:* `engineer-mobile-interactions`

**Stages:** Micro-interactions → Haptics

#### Role: Mobile Micro-Interaction Engineer

*Slug:* `mobile-micro-interaction-engineer` | *Stage:* Micro-interactions

*Domain:* Multi-touch gesture lifecycles, spring physics (stiffness, damping), 120Hz frame budgets, Reanimated v3 worklets, and native modifiers.

1. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Native mobile animation drivers, spring physics, and gesture recognizers.
2. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Fluid micro-interaction design and gesture state architecture.
3. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Fast mobile component implementation and state binding.
4. **`moonshotai/kimi-k2.7-code`** (262k ctx | $0.66 / $3.40) — Resilient mobile code generation.
5. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Mobile screen layout translation and asset mapping.

#### Role: Native Haptics Specialist

*Slug:* `native-haptics-specialist` | *Stage:* Haptics

*Domain:* Apple CoreHaptics (`.ahap` waveforms), Android `VibrationEffect` compositions, JSI hardware bridges, and tactile throttling.

1. **`mistralai/codestral-2508`** (256k ctx | $0.30 / $0.90) — Low-latency, precise hardware API bindings (Swift/Kotlin/C++).
2. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — High-throughput cross-platform haptic wrapper synthesizer.
3. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Fine-grained haptic feedback design and event mapping.
4. **`openai/gpt-5.4`** (1.05M ctx | $2.50 / $15.00) — Hardware sensor integration and battery lifecycle optimization.
5. **`openai/gpt-5.6-luna`** (1.05M ctx | $0.20 / $1.20) — Fast AHAP waveform JSON serialization.

---

### Workflow: Engineer Terminal Interfaces

*Slug:* `engineer-terminal-interfaces`

**Stages:** TUI layout → Event loop

#### Role: TUI Layout Engineer

*Slug:* `tui-layout-engineer` | *Stage:* TUI layout

*Domain:* Terminal emulation protocols (VT100, ANSI escape codes), TrueColor 24-bit rendering, alternate screens, and Ratatui constraint math.

1. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Pragmatic terminal systems code writer; adheres to strict TUI layout constraints.
2. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Complex ANSI rendering engines and terminal buffer diffing.
3. **`poolside/laguna-s-2.1`** (1.05M ctx | $0.09 / $0.18) — **Terminal-Bench 2.1 Specialist (70.2% score)**; 1M context at $0.09/M.
4. **`mistralai/devstral-2512`** (262k ctx | $0.40 / $2.00) — Clean Go (Bubbletea) and Rust (Ratatui) view/update implementations.
5. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — High-ergonomic terminal UX and keyboard navigation design.

#### Role: Terminal Event Loop Specialist

*Slug:* `terminal-event-loop-specialist` | *Stage:* Event loop

*Domain:* The Elm Architecture (TEA) in Bubbletea, Tokio event loops in Rust, raw mode stdin demuxing, Vim modal keymaps, and signal trapping.

1. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Flawless asynchronous event loop (Tokio/TEA) and POSIX termios handling.
2. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — Flagship CLI and sub-process navigation specialist; signal cleanup.
3. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — High-precision Rust/Go async event handling and state dispatch.
4. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Complex modal keybinding architectures (Vim/Emacs emulation).
5. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Clean command routing and keyboard navigation schemas.

---

### Workflow: Audit Visual Quality and Accessibility

*Slug:* `audit-visual-accessibility`

**Stages:** Visual QA → Accessibility

#### Role: Visual QA Polish Auditor

*Slug:* `visual-qa-polish-auditor` | *Stage:* Visual QA

*Domain:* High-resolution screenshot vs Figma spec comparisons, optical alignment, typography baseline grids, and concentric nested radii math.

1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — **Primary Automated Visual QA Gate**; rapid visual diff and layout regression detection.
2. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — High-precision design system aesthetic fidelity and alignment audit.
3. **`qwen/qwen3-vl-235b-a22b-instruct`** (262k ctx | $0.21 / $1.90) — Ultra-low-cost screenshot artifact and font rendering detector.
4. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Final qualitative visual polish and brand aesthetic signoff.
5. **`google/gemini-3.1-pro-preview`** (1.05M ctx | $2.00 / $12.00) — High-resolution pixel-perfect design-to-code comparison.

#### Role: Accessibility WCAG Specialist

*Slug:* `accessibility-wcag-specialist` | *Stage:* Accessibility

*Domain:* WCAG 2.2 AA/AAA success criteria, APCA contrast math, accessibility tree DOM verification, focus traps, and axe-core test automation.

1. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Gold standard for semantic HTML, ARIA tree hierarchies, and focus management.
2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Automated WCAG 2.2 color contrast and visual touch-target auditor.
3. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — High-throughput AST accessibility linter and automated ARIA fixer.
4. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Cognitive load and neurodiverse accessibility compliance evaluation.
5. **`openai/gpt-5.4`** (1.05M ctx | $2.50 / $15.00) — Screen-reader script simulation and compliance reporting.

---

## Media Production

*Slug:* `media-production`

**Overview:**

1. **Produce Generative Video:** Ideation $\to$ Scriptwriting $\to$ Shot Storyboarding $\to$ Prompt Synthesis $\to$ Video Generation $\to$ Visual Continuity QA $\to$ Audio/Foley Scoring $\to$ Final Render.
2. **Repurpose Long-Form Video:** Raw Media Ingestion (1–4h) $\to$ Semantic Hook Extraction $\to$ Active Speaker Tracking & Dynamic 9:16 Reframing $\to$ B-Roll Matching $\to$ Kinetic Typography & Packaging.
3. **Edit and Render Media:** Codec/Asset Ingestion $\to$ Rough-Cut Compilation $\to$ Deterministic NLE/FFmpeg Filtergraph Assembly $\to$ Color LUT/Loudness Normalization $\to$ Broadcast Compliance Gate.

---

### Workflow: Produce Generative Video

*Slug:* `produce-generative-video`

**Stages:** Script and storyboard → Prompt engineering → Visual continuity → Video QA → Audio-visual compositing

#### Role: Scriptwriter Storyboard Director

*Slug:* `scriptwriter-storyboard-director` | *Stage:* Script and storyboard

*Domain:* Cinematic screenplays, narrative arcs, shot descriptions, scene tempos, and blocking instructions.

1. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Frontier narrative depth, emotional nuance, and long-horizon multi-act pacing across 1M context.
2. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Highly responsive, steerable screenplay formatting (Fountain/Final Draft XML) at 1/3rd the operating cost.
3. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Punchy dialogue cadence, contemporary cultural fluency, and unconventional creative treatments.
4. **`mistralai/mistral-large-2512`** (262k ctx | $0.50 / $1.50) — Multilingual narrative drafting and shot list breakdown at exceptional unit economics.
5. **`openai/gpt-6-astra-pro`** (1.05M ctx | $10.00 / $50.00) — Complex multi-character worldbuilding and director-ready blocking notes.

#### Role: Video Prompt Engineer

*Slug:* `video-prompt-engineer` | *Stage:* Prompt engineering

*Domain:* Deconstructing storyboards into diffusion/autoregressive prompts (framing, camera vectors, lighting, negative constraints).

1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Fast spatial-visual comprehension; converts visual concepts into diffusion parameters at high throughput.
2. **`bytedance-seed/seed-2-1-turbo`** (262k ctx | $0.50 / $2.50) — Native diffusion tuning; outputs compact, weighted prompts with motion velocity vectors.
3. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Complex camera motion prompting (e.g., *dolly zoom with 45° pan*) without semantic drift.
4. **`qwen/qwen3.8-max-0902`** (1M ctx | $2.00 / $6.00) — 2.4T multimodal MoE tuned across global generative video backends (Wan, Kling, Sora, Runway).
5. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $2.00 / $10.00) — Strict schema compliance for programmatic video generation APIs.

#### Role: Visual Continuity Engine

*Slug:* `visual-continuity-engine` | *Stage:* Visual continuity

*Domain:* Compares generated frames against reference sheets to prevent facial, wardrobe, lighting, and environmental drift.

1. **`google/gemini-3.1-pro-preview`** (1.05M ctx | $2.00 / $12.00) — Native multi-image/video reasoning; cross-references character turnarounds across a 1M token context.
2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Real-time frame-by-frame similarity scoring and spatial discrepancy detection.
3. **`qwen/qwen3.8-max-0902`** (1M ctx | $2.00 / $6.00) — Fine-grained texture, logo, and wardrobe consistency tracking.
4. **`z-ai/glm-5.3-flash`** (1.31M ctx | $0.075 / $0.25) — Ultra-low-cost screening gate across high-volume rendering batches.
5. **`openai/gpt-5.4-image-2`** (272k ctx | $8.00 / $15.00) — *Image Output Engine*; generates in-painted corrective reference keyframes.

#### Role: Video QA Defect Inspector

*Slug:* `video-qa-defect-inspector` | *Stage:* Video QA

*Domain:* Scans raw video renders for frame flicker, non-physical deformations, temporal morphing, and anatomical glitches.

1. **`z-ai/glm-5.3-flash`** (1.31M ctx | $0.075 / $0.25) — **Tier-0 Primary Ingestion Gate**; 1.31M native video context at $0.075/M prompt.
2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Fast, reliable defect classification and timestamp annotation.
3. **`google/gemini-3.1-pro-preview`** (1.05M ctx | $2.00 / $12.00) — Escalation arbiter for subtle temporal and physical artifact validation.
4. **`qwen/qwen3.8-max-0902`** (1M ctx | $2.00 / $6.00) — Evaluates physics continuity (fluid dynamics, shadows, reflections).
5. **`bytedance-seed/seed-2-1-turbo`** (262k ctx | $0.50 / $2.50) — Emits structured defect reports categorized by error severity.

#### Role: Audio-Visual Compositor

*Slug:* `audio-visual-compositor` | *Stage:* Audio-visual compositing

*Domain:* Analyzes video hit points, generates music briefs, places Foley effects, and aligns audio with visual cues.

1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Native multimodal engine (video + audio + text); detects hit points and sync offsets.
2. **`google/lyria-3-pro-preview`** (1.05M ctx | $0.08 / song) — Generates high-fidelity 48kHz audio and scoring directly from visual prompts.
3. **`openai/gpt-audio`** (128k ctx | $2.50 / $10.00) — Native speech and audio synthesis; cadence and vocal inflection alignment.
4. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Authors time-indexed EDL/FCPXML sound design cue sheets.
5. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $2.00 / $10.00) — Orchestrates multi-channel frequency and ducking parameter hierarchies.

---

### Workflow: Repurpose Long-Form Video

*Slug:* `repurpose-long-form-video`

**Stages:** Highlight extraction → Reframing and B-roll → Kinetic typography

#### Role: Highlight Hook Extractor

*Slug:* `highlight-hook-extractor` | *Stage:* Highlight extraction

*Domain:* Ingests 1–4 hour recordings, identifying viral hooks, high-retention segments, and standalone topics.

1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Scans multi-hour video/audio streams; extracts timestamped clips at sub-$0.50 cost per hour.
2. **`z-ai/glm-5.3-flash`** (1.31M ctx | $0.075 / $0.25) — Massive 1.31M context for bulk background indexing of livestreams.
3. **`google/gemini-3.1-pro-preview`** (1.05M ctx | $2.00 / $12.00) — Deep semantic extraction for complex lectures and technical panels.
4. **`qwen/qwen3.8-max-0902`** (1M ctx | $2.00 / $6.00) — Correlates audience physical cues (laughter, applause) with audio punchlines.
5. **`moonshotai/kimi-k3`** (1.05M ctx | $3.00 / $15.00) — Long-horizon needle-in-a-haystack topic retrieval across long discussions.

#### Role: Reframing B-Roll Matcher

*Slug:* `reframing-broll-matcher` | *Stage:* Reframing and B-roll

*Domain:* Tracks active speakers for 16:9 to 9:16 dynamic cropping and generates semantic B-roll search queries.

1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Computes normalized bounding-box coordinates for smooth virtual camera pans.
2. **`qwen/qwen3.8-flash`** (1M ctx | $0.15 / $0.47) — Real-time contextual B-roll keyword and stock asset search query generation.
3. **`xiaomi/mimo-v2.5`** (1.05M ctx | $0.14 / $0.28) — Native omnimodal video model with spatial frame perception.
4. **`bytedance-seed/seed-2-1-turbo`** (262k ctx | $0.50 / $2.50) — Visual tracking and keyframe coordinate mapping.
5. **`amazon/nova-2-lite-v1`** (1M ctx | $0.30 / $2.50) — Low-cost video indexing and visual scene classification.

#### Role: Kinetic Typographer

*Slug:* `kinetic-typographer` | *Stage:* Kinetic typography

*Domain:* Formats word-level subtitle timings, selects kinetic text styling/colors, and drafts click-through titles.

1. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — High cultural fluency; generates punchy titles, descriptions, and hashtags matching platform trends.
2. **`mistralai/mistral-large-2512`** (262k ctx | $0.50 / $1.50) — Cost-effective structured JSON/ASS subtitle styling and localization.
3. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Polished typography formatting, emphasis palettes, and social pacing.
4. **`qwen/qwen3.8-flash`** (1M ctx | $0.15 / $0.47) — Real-time formatting of raw Whisper transcripts into styled subtitle chunks.
5. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $2.00 / $10.00) — Precise timing verification and deterministic subtitle schema alignment.

---

### Workflow: Edit and Render Media

*Slug:* `edit-render-media`

**Stages:** Timeline assembly → Broadcast compliance

#### Role: Timeline Assembly Engineer

*Slug:* `timeline-assembly-engineer` | *Stage:* Timeline assembly

*Domain:* Emits executable FFmpeg filtergraphs, OpenTimelineIO (OTIO) JSON, and Remotion/DaVinci Resolve Python scripts.

1. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Frontier code generation for complex multi-input FFmpeg filtergraphs and OTIO structures.
2. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — 1M-context Remotion/React-video and Python automation specialist at low cost.
3. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Multi-track timecode arithmetic, drop-frame calculations, and nested timeline logic.
4. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Inspects raw asset metadata directly and constructs execution CLI commands.
5. **`mistralai/devstral-2512`** (262k ctx | $0.40 / $2.00) — Lightweight Bash and Python timeline glue scripts.

#### Role: Broadcast Compliance Gatekeeper

*Slug:* `broadcast-compliance-gatekeeper` | *Stage:* Broadcast compliance

*Domain:* Verifies broadcast safe zones, subtitle synchronization, audio loudness (EBU R128), and photosensitive epilepsy (PSE) safety.

1. **`google/gemini-3.1-pro-preview`** (1.05M ctx | $2.00 / $12.00) — Multi-modal audit across 1M tokens; validates title safety, strobe risks, and branding.
2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Fast automated gatekeeper in rendering CI/CD; catches black frames and desync.
3. **`meta/muse-spark-1.3`** (1.05M ctx | $1.25 / $4.25) — Native audio/video inspection of loudness dynamics and visual pacing.
4. **`qwen/qwen3.8-max-0902`** (1M ctx | $2.00 / $6.00) — High spatial resolution checks for color clipping and compression banding.
5. **`z-ai/glm-5.3-flash`** (1.31M ctx | $0.075 / $0.25) — High-throughput preliminary broadcast screening.

---

## Data & Analytics

*Slug:* `data-analytics`

**Overview:**

1. **Query Business Intelligence:** Metric Resolution $\to$ SQL Generation $\to$ AST Optimization $\to$ Warehouse Execution $\to$ Declarative Dashboard Formatting.
2. **Analyze Dataset:** Data Hygiene $\to$ Hypothesis Formulation $\to$ Vectorized Python (Polars/Pandas) $\to$ Sandbox Execution $\to$ Causal Inference.
3. **Extract and Audit Documents:** Massive 10-K Ingestion $\to$ Table/Chart Extraction $\to$ High-Precision Mathematical Auditing $\to$ Executive Variance Memo.

---

### Workflow: Query Business Intelligence

*Slug:* `query-business-intelligence`

**Stages:** Text-to-SQL → Query optimization → Dashboard formatting

#### Role: Text-to-SQL Architect

*Slug:* `text-to-sql-architect` | *Stage:* Text-to-SQL

*Domain:* Translates natural language questions into performant SQL queries (CTEs, window functions) across enterprise warehouses.

1. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Complex nested SQL queries, CTEs, and window functions.
2. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — Ingests 1,000-table enterprise data dictionaries in 1M context.
3. **`deepseek/deepseek-v4-pro`** (1.05M ctx | $0.63 / $1.25) — High-throughput Text-to-SQL generation.
4. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Ambiguous business metric semantic translation.
5. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Ultra-fast schema resolution and join mapping.

#### Role: SQL Optimizer Validator

*Slug:* `sql-optimizer-validator` | *Stage:* Query optimization

*Domain:* Inspects `EXPLAIN ANALYZE` query plans, eliminates Cartesian products, and applies dialect-specific index optimizations.

1. **`openai/o3`** (200k ctx | $2.00 / $8.00) — Query execution plan optimization and index utilization logic.
2. **`deepseek/deepseek-r1-0528`** (164k ctx | $0.50 / $2.15) — High-efficiency algorithmic query rewriting.
3. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Database engine-specific syntax correction (Postgres/Snowflake/BigQuery).
4. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — Cost-efficient automated query correction loop.
5. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Enterprise data governance and query compliance.

#### Role: BI Dashboard Formatter

*Slug:* `bi-dashboard-formatter` | *Stage:* Dashboard formatting

*Domain:* Transforms SQL result tables into declarative chart specifications (Vega-Lite, ECharts) with executive commentary.

1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Visual chart layout and structured Vega-Lite/ECharts JSON.
2. **`qwen/qwen3.8-flash`** (1M ctx | $0.15 / $0.47) — Ultra-fast, low-cost dashboard widget JSON generation.
3. **`z-ai/glm-5.3-flash`** (1.31M ctx | $0.075 / $0.25) — High-throughput dashboard data series formatting.
4. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Executive visual narrative and metric layout design.
5. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $2.00 / $10.00) — Strict schema compliance for BI embedded frameworks.

---

### Workflow: Analyze Dataset

*Slug:* `analyze-dataset`

**Stages:** Code synthesis → Causal reasoning

#### Role: Data Science Code Synthesizer

*Slug:* `data-science-code-synthesizer` | *Stage:* Code synthesis

*Domain:* Writes vectorized, performant Python code for feature engineering, statistical modeling, and ML pipelines.

1. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Complex numerical analysis, PyTorch/Scikit pipelines.
2. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — High-capacity Python data science code generator with 1M context.
3. **`mistralai/codestral-2508`** (256k ctx | $0.30 / $0.90) — Extremely low-cost Polars/Pandas data pipeline implementation.
4. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Complex econometric and Bayesian model formulation.
5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Large-scale data ingestion and transformation scripts.

#### Role: Statistical Causal Reasoner

*Slug:* `statistical-causal-reasoner` | *Stage:* Causal reasoning

*Domain:* Evaluates p-values, separates correlation from causation, controls for confounders, and diagnoses distribution drift.

1. **`openai/o3`** (200k ctx | $2.00 / $8.00) — Advanced causal inference, covariate adjustment, and distribution shift reasoning.
2. **`deepseek/deepseek-r1-0528`** (164k ctx | $0.50 / $2.15) — High-efficiency statistical hypothesis testing and anomaly deduction.
3. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Interpreting causal models for business decision-makers.
4. **`google/gemini-3.1-pro-preview`** (1.05M ctx | $2.00 / $12.00) — Ingestion of multimodal data distribution plots and correlations.
5. **`openai/gpt-6-astra-pro`** (1.05M ctx | $10.00 / $50.00) — Enterprise-scale econometric simulation.

---

### Workflow: Extract and Audit Documents

*Slug:* `extract-audit-documents`

**Stages:** Dossier extraction → Variance audit

#### Role: Financial Dossier Extractor

*Slug:* `financial-dossier-extractor` | *Stage:* Dossier extraction

*Domain:* Ingests 100k–1M+ token SEC 10-K/10-Q filings and investor decks; extracts dense tables and footnotes without OCR degradation.

1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Premier unit economics for 1M-token PDF table and chart extraction.
2. **`google/gemini-3.1-pro-preview`** (1.05M ctx | $2.00 / $12.00) — Footnote reconciliation and complex accounting table extraction.
3. **`qwen/qwen3.8-max-0902`** (1M ctx | $2.00 / $6.00) — Robust bilingual financial reporting and balance sheet analysis.
4. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Qualitative MD&A disclosure risk synthesis.
5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Low-cost long-form structured financial data normalization.

#### Role: Financial Variance Auditor

*Slug:* `financial-variance-auditor` | *Stage:* Variance audit

*Domain:* Reconciles statement equations, verifies debt covenants, and recalculates financial ratios with zero arithmetic error.

1. **`openai/o3`** (200k ctx | $2.00 / $8.00) — Zero-hallucination arithmetic audit and ledger balance reconciliation.
2. **`deepseek/deepseek-r1-0528`** (164k ctx | $0.50 / $2.15) — High-speed variance detection and formula recalculation.
3. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Full financial audit report synthesis and regulatory commentary.
4. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $2.00 / $10.00) — Audit compliance verification against accounting standards.
5. **`openai/gpt-6-astra-pro`** (1.05M ctx | $10.00 / $50.00) — Complex tax and holding company consolidation.

---

## Research & Strategy

*Slug:* `research-strategy`

**Overview:**

1. **Prepare Decision Brief:** Status Ingestion $\to$ Signal Extraction $\to$ Scenario Trade-Off Modeling $\to$ C-Suite Ghostwriting $\to$ Tone Calibration.
2. **Structure Negotiations:** Contract/Ticket Ingestion $\to$ Leverage (BATNA) Modeling $\to$ Counter-Proposal Structuring.

---

### Workflow: Prepare Decision Brief

*Slug:* `prepare-decision-brief`

**Stages:** Intelligence synthesis → Scenario modeling → Competitive research

#### Role: Executive Intelligence Synthesizer

*Slug:* `executive-intelligence-synthesizer` | *Stage:* Intelligence synthesis

*Domain:* Distills messy operational data into dense, crisp, jargon-free executive narratives and board-level memos.

1. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Gold standard in executive prose; authoritative C-suite memos with zero fluff.
2. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Rapid daily drafting of executive summaries and decision briefs.
3. **`writer/palmyra-x5`** (1.04M ctx | $0.60 / $6.00) — Enterprise business corpus trained; low-cost long-form synthesis.
4. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Direct, unvarnished operational status reports.
5. **`openai/gpt-6-astra`** (1.05M ctx | $10.00 / $50.00) — High-stakes multi-entity corporate restructuring.

#### Role: Strategic Scenario Modeler

*Slug:* `strategic-scenario-modeler` | *Stage:* Scenario modeling

*Domain:* Analyzes capital allocation trade-offs, competitive moats (7 Powers), and down-side risk matrices.

1. **`openai/o3`** (200k ctx | $2.00 / $8.00) — Game-theoretic modeling of competitor reactions and pricing power.
2. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Evaluates systemic industry dynamics and platform lock-in effects.
3. **`deepseek/deepseek-r1-0528`** (164k ctx | $0.50 / $2.15) — Pure chain-of-thought mathematical sensitivity analysis.
4. **`qwen/qwen3.8-max-0902`** (1M ctx | $2.00 / $6.00) — Macroeconomic supply chain disruption analysis across 1M context.
5. **`openai/gpt-6-astra`** (1.05M ctx | $10.00 / $50.00) — Multi-variable enterprise stress testing.

#### Role: Competitive Web Researcher

*Slug:* `competitive-web-researcher` | *Stage:* Competitive research

*Domain:* Scans public databases, competitor releases, earnings calls, and patent filings to build factual dossiers.

1. **`perplexity/sonar-deep-research`** (128k ctx | $2.00 / $8.00) — Autonomous multi-step deep-web market research with live citations.
2. **`x-ai/grok-4.20-multi-agent`** (2M ctx | $1.25 / $2.50) — Real-time social, news, and market intelligence across 2M context.
3. **`perplexity/sonar-pro-search`** (200k ctx | $3.00 / $15.00) — Rapid multi-source competitive landscape synthesis.
4. **`google/gemini-3.1-pro-preview`** (1.05M ctx | $2.00 / $12.00) — Ingests competitor investor slide decks, financials, and whitepapers.
5. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Continuous automated monitoring of competitor site changes.

---

### Workflow: Structure Negotiations

*Slug:* `structure-negotiations`

**Stages:** Deal strategy

#### Role: Negotiation Deal Strategist

*Slug:* `negotiation-deal-strategist` | *Stage:* Deal strategy

*Domain:* Models counterparty BATNA, calculates walk-away thresholds, and structures tiered commercial counter-proposals.

1. **`openai/o3`** (200k ctx | $2.00 / $8.00) — Game-theoretic concession modeling and BATNA optimization.
2. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Psychological leverage, interpersonal dynamics, and counter-framing.
3. **`deepseek/deepseek-r1-0528`** (164k ctx | $0.50 / $2.15) — Mathematical deal structure and incentive alignment analysis.
4. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — Term sheet covenant structuring.
5. **`openai/gpt-6-astra`** (1.05M ctx | $10.00 / $50.00) — Complex multi-party merger and joint venture dynamics.

---

## Business Operations

*Slug:* `business-operations`

**Overview:**

1. **Automate Tasks:** Multi-Channel Ingestion $\to$ Urgency Triage $\to$ Action Item Extraction $\to$ Multi-Tool API Dispatch $\to$ Daily Briefing.
2. **Handle Customer Escalations:** Ticket Ingestion $\to$ De-escalation Comms.
3. **Review Contracts:** Contract Ingestion $\to$ SLA/Legal Risk Audit $\to$ Redline.

---

### Workflow: Automate Tasks

*Slug:* `automate-tasks`

**Stages:** Tool dispatch → Inbox triage → Executive briefing

#### Role: Workflow Tool Dispatcher

*Slug:* `workflow-tool-dispatcher` | *Stage:* Tool dispatch

*Domain:* Invokes external APIs (calendar, email, CRM), coordinates webhooks, and manages background task pipelines.

1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Sub-second deterministic tool dispatch with zero argument hallucination.
2. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — Complex multi-step tool dependency chaining and error recovery.
3. **`qwen/qwen3.6-flash`** (1M ctx | $0.1875 / $1.125) — High-throughput sub-second personal action router.
4. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Ambiguous user intent disambiguation.
5. **`google/gemini-3.1-pro-preview-customtools`** (1.05M ctx | $2.00 / $12.00) — Multi-agent tool orchestration.

#### Role: Inbox Communication Triager

*Slug:* `inbox-communication-triager` | *Stage:* Inbox triage

*Domain:* Classifies incoming emails, extracts action items, prioritizes blockers, and drafts context-aware replies.

1. **`deepseek/deepseek-v4-flash-0731`** (1.31M ctx | $0.05 / $0.10) — High-volume Tier-0 spam/priority email triage at $0.05/M.
2. **`openai/gpt-5.6-luna`** (1.05M ctx | $0.20 / $1.20) — Fast structured JSON action item extractor from emails.
3. **`google/gemini-3.5-flash-lite`** (1.05M ctx | $0.30 / $2.50) — Multimodal attachment and PDF receipt processing.
4. **`mistralai/mistral-small-3.2-24b-instruct`** (131k ctx | $0.075 / $0.20) — Private, fast thread summarization.
5. **`anthropic/claude-haiku-4.5`** (200k ctx | $1.00 / $5.00) — Escalation tier for sensitive VIP relationship messages.

#### Role: Executive Briefing Concierge

*Slug:* `executive-briefing-concierge` | *Stage:* Executive briefing

*Domain:* Aggregates meeting transcripts, daily agendas, and commitments into a prioritized morning briefing.

1. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Optimal balance of executive briefing tone, elegance, and cost.
2. **`upstage/solar-pro4`** (524k ctx | $0.03 / $0.12) — High-efficiency calendar, news, and slack backlog summarizer.
3. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — High-stakes executive priority briefing and talking points.
4. **`x-ai/grok-4.3`** (1M ctx | $1.25 / $2.50) — Fast morning social and industry pulse digest.
5. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Multimodal briefing compiling audio voicemails and visual slide decks.

---

### Workflow: Handle Customer Escalations

*Slug:* `handle-customer-escalations`

**Stages:** De-escalation communication

#### Role: Customer De-escalation Communicator

*Slug:* `customer-deescalation-communicator` | *Stage:* De-escalation communication

*Domain:* Authors empathetic, authoritative, legally sound communications during outages, breaches, or disputes.

1. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Masterclass in diplomatic de-escalation, emotional intelligence, and trust restoration.
2. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Fast turnaround for high-pressure incident customer comms.
3. **`openai/gpt-6-astra`** (1.05M ctx | $10.00 / $50.00) — High-stakes enterprise SLA breach statements.
4. **`writer/palmyra-x5`** (1.04M ctx | $0.60 / $6.00) — Corporate PR and compliance-safe messaging.
5. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Direct, transparent post-mortem communication.

---

### Workflow: Review Contracts

*Slug:* `review-contracts`

**Stages:** Contract redline

#### Role: Contract Redline Auditor

*Slug:* `contract-redline-auditor` | *Stage:* Contract redline

*Domain:* Audits 100-page MSAs, DPAs, and term sheets; flags aggressive indemnification, liability caps, and non-standard SLAs.

1. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Enterprise legal risk auditor; non-negotiable liability clause verification.
2. **`writer/palmyra-x5`** (1.04M ctx | $0.60 / $6.00) — Purpose-built contract language and indemnification clause analyzer.
3. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Contextual commercial terms and dispute resolution redlines.
4. **`google/gemini-3.1-pro-preview`** (1.05M ctx | $2.00 / $12.00) — Long 1M-token cross-contract reconciliation.
5. **`openai/gpt-6-astra`** (1.05M ctx | $10.00 / $50.00) — Cross-border regulatory compliance verification.

---

## Security & Reliability

*Slug:* `security-reliability`

**Overview:**

1. **Investigate Incident:** Telemetry filtering $\to$ Incident triage $\to$ Forensic analysis $\to$ Reproduction $\to$ Debrief. Stages are optional per incident; there is no mandatory five-model fanout.
2. **Patch Vulnerability:** Taint Flow Analysis $\to$ Defensive Patching $\to$ Adversarial Mutation Verification.

---

### Workflow: Investigate Incident

*Slug:* `investigate-incident`

**Stages:** Telemetry filtering → Incident triage → Forensic analysis → Reproduction → Debrief (optional per incident; no mandatory five-model fanout)

#### Role: Telemetry Stream Classifier

*Slug:* `telemetry-stream-classifier` | *Stage:* Telemetry filtering

*Domain:* Ingests high-velocity JSON logs, OpenTelemetry traces, and time-series metrics at rock-bottom token cost.

1. **`qwen/qwen3.7-flash`** (1M ctx | $0.03 / $0.13) — **Tier-0 Primary Streaming Gate**; 1M context at $0.03/M prompt.
2. **`deepseek/deepseek-v4-flash-0731`** (1.31M ctx | $0.05 / $0.10) — High-throughput anomaly classification in 1.31M context.
3. **`z-ai/glm-5.3-flash`** (1.31M ctx | $0.075 / $0.25) — High-speed log aggregation and pattern recognition.
4. **`google/gemini-3.1-flash-lite`** (1.05M ctx | $0.25 / $1.50) — Multimodal metric graph and trace ingestion.
5. **`amazon/nova-2-lite-v1`** (1M ctx | $0.30 / $2.50) — Resilient secondary cloud telemetry classifier.

#### Role: Incident Triage Investigator

*Slug:* `incident-triage-investigator` | *Stage:* Incident triage

*Domain:* Consumes raw multi-gigabyte log dumps, APM spans, and core dumps; filters noise and isolates fault boundaries.

1. **`deepseek/deepseek-v4-flash-0731`** (1.31M ctx | $0.05 / $0.10) — **Tier-0 Primary Ingestion Gate**; 1.31M context at $0.05/M prompt.
2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Multimodal crash analysis (logs + APM graphs + error screenshots).
3. **`openai/gpt-5.6-luna`** (1.05M ctx | $0.20 / $1.20) — Fast structured JSON classification of stack traces across 1M tokens.
4. **`x-ai/grok-4.20`** (2M ctx | $1.25 / $2.50) — 2M context window for cross-service distributed tracing scans.
5. **`qwen/qwen3.8-flash`** (1M ctx | $0.15 / $0.47) — High-throughput multilingual log scanner for multi-cluster events.

#### Role: Forensic Causal Analyst

*Slug:* `forensic-causal-analyst` | *Stage:* Forensic analysis

*Domain:* Analyzes concurrent thread traces, memory corruptions, and race conditions to isolate line-level defects.

1. **`openai/o3`** (200k ctx | $2.00 / $8.00) — Frontier test-time reasoning on distilled call graphs; uncovers edge-case race conditions.
2. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Deep forensic reasoning across complex asynchronous lifecycles in 1M context.
3. **`deepseek/deepseek-r1-0528`** (164k ctx | $0.50 / $2.15) — Open chain-of-thought mathematical and logical bug deduction.
4. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Multi-repository diagnostic investigations for systemic architectural failures.
5. **`openai/gpt-6-astra`** (1.05M ctx | $10.00 / $50.00) — Long-context multi-system causal graph reconstruction.

#### Role: Repro PoC Synthesizer

*Slug:* `repro-poc-synthesizer` | *Stage:* Reproduction

*Domain:* Translates trace observations into isolated, runnable reproduction unit/integration tests and scripts.

1. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Fast, repo-accurate test writer; minimal boilerplate without hallucinating dependencies.
2. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — Ingests full test frameworks; generates parameterized reproduction suites.
3. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Specialized mocking of network calls, clocks, and internal database states.
4. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Parameterized property-based regression test harnesses (Hypothesis/QuickCheck).
5. **`mistralai/devstral-2512`** (262k ctx | $0.40 / $2.00) — Concise, executable unit-test reproductions.

#### Role: Incident Debrief Synthesizer

*Slug:* `incident-debrief-synthesizer` | *Stage:* Debrief

*Domain:* Traces microservice failure cascades, determines primary failure mechanisms, and authors post-mortems.

1. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Unmatched incident debrief authoring, timeline reconstruction, and learnings.
2. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — High-velocity technical post-mortem report writer.
3. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Root cause extraction without corporate sugarcoating.
4. **`google/gemini-3.8-flash`** (1.05M ctx | $0.75 / $3.75) — Visual timeline generation combining metric spikes and log traces.
5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Infrastructure remediations and preventive patch specification.

---

### Workflow: Patch Vulnerability

*Slug:* `patch-vulnerability`

**Stages:** Defensive patching → Adversarial audit

#### Role: Defensive Patch Hardening Engineer

*Slug:* `defensive-patch-hardening-engineer` | *Stage:* Defensive patching

*Domain:* Synthesizes fail-closed security fixes (parameterized queries, sanitization, constant-time comparisons).

1. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Surgical security patch generator using secure coding primitives.
2. **`x-ai/grok-4.6`** (500k ctx | $2.00 / $6.00) — Rapid patch application that passes `npm run verify` cleanly.
3. **`qwen/qwen3-coder-plus`** (1M ctx | $0.65 / $3.25) — High-throughput defensive refactoring across large repos.
4. **`anthropic/claude-sonnet-5`** (1M ctx | $2.00 / $10.00) — Applies defensive validation without breaking valid user flows.
5. **`bytedance-seed/seed-2.0-code`** (262k ctx | $0.50 / $3.00) — Web and client-side sanitization hardening.

#### Role: Adversarial Security Mutation Auditor

*Slug:* `adversarial-security-mutation-auditor` | *Stage:* Adversarial audit

*Domain:* Authors exploit PoCs, executes mutation testing, and ensures patches introduce zero bypass vectors.

1. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $2.00 / $10.00) — Pro reasoning mode; probes candidate fixes with adversarial bypass payloads.
2. **`openai/o3`** (200k ctx | $2.00 / $8.00) — Fuzzing parameter generation and mathematical boundary mutation testing.
3. **`anthropic/claude-opus-5`** (1M ctx | $5.00 / $25.00) — Unforgiving security critic; catches incomplete sanitization and secondary injections.
4. **`nvidia/nemotron-3-ultra-550b-a55b`** (262k ctx | $0.625 / $3.125) — 550B MoE frontier open reasoning model for vulnerability search.
5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.58 / $1.74) — Full-codebase taint tracking and dataflow analysis.

---

## Historical Navigation

This table is historical navigation from the 2026-09-06 heading numbers to the current documentation slugs. It is not a runtime alias lane.

| Old heading | New workflow slug | Old roles |
|---|---|---|
| Workflow 1.1: Generative AI Video Production Pipeline | `produce-generative-video` | 1.1.1–1.1.5 |
| Workflow 1.2: Automated Long-to-Shorts Content Repurposing Pipeline | `repurpose-long-form-video` | 1.2.1–1.2.3 |
| Workflow 1.3: Programmatic Post-Production & Tool-Calling Assembly Pipeline | `edit-render-media` | 1.3.1–1.3.2 |
| Workflow 2.1: Greenfield Software Engineering & Monorepo Scaffolding | `build-feature` | 2.1.1–2.1.6 |
| Workflow 3.1: Production Incident RCA & Crash Diagnostics | `investigate-incident` | 3.1.1–3.1.3 |
| Workflow 3.2: Precision Code Refactoring & Regression Repair | `refactor-repair-regressions` | 3.2.1–3.2.3 |
| Workflow 3.3: Flaky Test Remediation & Vulnerability Patching | (split; see role rows) | 3.3.1–3.3.3 |
| Role 3.3.1: Concurrency & Flakiness Detective | `stabilize-flaky-tests` | 3.3.1 |
| Role 3.3.2: Defensive Patch & Hardening Engineer | `patch-vulnerability` | 3.3.2 |
| Role 3.3.3: Adversarial Security & Mutation Auditor | `patch-vulnerability` | 3.3.3 |
| Workflow 4.1: Executive Decision Support & Strategic Planning | `prepare-decision-brief` | 4.1.1–4.1.3 |
| Workflow 4.2: Autonomous Personal Productivity & Task Automation | `automate-tasks` | 4.2.1–4.2.3 |
| Workflow 4.3: High-Stakes Customer Escalation, Negotiation & Contracts | (split; see role rows) | 4.3.1–4.3.3 |
| Role 4.3.1: Contract Redline & Commercial Terms Auditor | `review-contracts` | 4.3.1 |
| Role 4.3.2: Crisis & Customer De-escalation Communicator | `handle-customer-escalations` | 4.3.2 |
| Role 4.3.3: Negotiation Leverage & Deal Structuring Strategist | `structure-negotiations` | 4.3.3 |
| Workflow 5.1: Distributed Systems Design & Technical RFC Formulation | `design-software-system` | 5.1.1–5.1.6 |
| Workflow 6.1: Enterprise Business Intelligence & Text-to-SQL Pipeline | `query-business-intelligence` | 6.1.1–6.1.3 |
| Workflow 6.2: Automated Data Science & Statistical Modeling | `analyze-dataset` | 6.2.1–6.2.2 |
| Workflow 6.3: Multimodal Financial & Operational Document Intelligence | `extract-audit-documents` | 6.3.1–6.3.2 |
| Workflow 6.4: Real-Time Operational Telemetry & Distributed Systems RCA | `investigate-incident` | 6.4.1–6.4.2 |
| Workflow 7.1: Design System Architecture & Multi-Platform Component Engineering | `build-design-system` | 7.1.1–7.1.2 |
| Workflow 7.2: Mobile-First Interactive UX & Gesture/Haptic Engineering | `engineer-mobile-interactions` | 7.2.1–7.2.2 |
| Workflow 7.3: Terminal User Interface (TUI) & Rich CLI Experience Engineering | `engineer-terminal-interfaces` | 7.3.1–7.3.2 |
| Workflow 7.4: Multimodal Visual Design QA, Accessibility (a11y) & Polish Audit | `audit-visual-accessibility` | 7.4.1–7.4.2 |

Role rotation, admission, and dispatch policy live in AGENTS.md and Tracking, not in this guide.