@kontextmind/kxm 0.7.208 → 0.7.210

package/.claude-plugin/marketplace.json

@@ -11,7 +11,7 @@
       "name": "kxm",
       "source": "./plugins/kxm",
       "description": "Durable workflows, peer agents, and kxm tui",
-      "version": "0.7.208",
+      "version": "0.7.210",
       "category": "development",
       "tags": ["kxm", "multi-agent", "workflows", "mcp"]
     }

package/CHANGELOG.md

@@ -225,6 +225,26 @@ All notable user-facing changes are documented here. The project follows [Semant
 
 ### Changed
 
+- **Workflow catalog prices were re-verified against OpenRouter on 2026-10-29.**
+  Candidate context windows and list prices in
+  [the workflow catalog](docs/reference/workflow-catalog.md) match
+  `https://openrouter.ai/api/v1/models` from that day. `openai/o3-mini-high`
+  and `kwaipilot/kat-coder-pro-v2.5` are gone. The semantic equivalence
+  verifier ranks `openai/gpt-6.1-sol` first, the CLI critic, then
+  `openai/gpt-5.6-sol`. The old Kat Coder slots name `qwen/qwen3-coder`.
+  Catalog roles that named `x-ai/grok-4.6` now name `x-ai/grok-4.7`. The
+  writer roster is unchanged. Guided setup follows the software-engineering
+  lists, so a signed-in `grok` harness is admitted as `xai/grok-4.7`, and
+  the equivalence verifier can select `gpt-6.1-sol`. `.kxm/prices.yaml` is
+  dated 2026-10-29. `z-ai/glm-5.3-flash` is still $0.15 input and $0.50
+  output per 1M tokens (cache read $0.03), context 1,048,576. That list
+  price is volatile. `zai-coding-cn/glm-5.3` follows the public list price,
+  $0.0699 input and $1.40 output (cache read $0.0689).
+  `qwen-token-plan/deepseek-v4.1-flash` follows the public list price for
+  its alias, $0.045 input and $0.50 output (cache read $0.006). The weekly
+  refresh still does not rewrite those two non-OpenRouter rows. Inventory
+  `standard` rates were refreshed the same day. GLM 4.6 remains $0.50 / $2.00
+  at 204,800 context.
 - **OpenRouter GLM 5.3 Flash is $0.15 / $0.50 again.**
   Checked 2026-10-21 against `https://openrouter.ai/api/v1/models`
   (`z-ai/glm-5.3-flash`: prompt `0.00000015`, completion `0.0000005`,
@@ -234,8 +254,8 @@ All notable user-facing changes are documented here. The project follows [Semant
   `zai-coding-cn/glm-5.3-flash` keeps its own tier at $0.15 / $0.50 (cache
   read $0.03). This list price is volatile
   and has flipped between those two tiers for several days. Trust the weekly
-  OpenRouter refresh over the static row. The inventory `standard` rates for
-  `z-ai/glm-5.3-flash` match. `.kxm/prices.yaml` is dated 2026-10-21.
+  OpenRouter   refresh over the static row. The inventory `standard` rates for
+  `z-ai/glm-5.3-flash` match. The 2026-10-21 catalog recorded that move.
 - **`kxm vision assert` defaults to the admitted OpenRouter GLM route.**
   The built-in route is `openrouter/z-ai/glm-5.3-flash`, invoked as
   `pi --model openrouter/z-ai/glm-5.3-flash`. The old default
@@ -250,8 +270,8 @@ All notable user-facing changes are documented here. The project follows [Semant
   `xai/grok-4.7`, alias `grok-default`). The writer roster, the built-in
   coder and browser modes, the general subagent default, the Pi producer
   default, and the `kxm routing benchmark` arm list use the same model.
-  Guide candidates and selector examples may still show `grok-4.6`.
-  `.kxm/prices.yaml` keeps both price rows.
+  Selector examples may still show `grok-4.6`. Guide candidates name
+  `grok-4.7`. `.kxm/prices.yaml` keeps both price rows.
 
 - **The planner fallback is Claude Sonnet 5.5.** `claude-fable` stays the
   planner primary. `claude-claude-sonnet-5-5` (`claude` / `claude-sonnet-5.5`,

package/docs/reference/workflow-catalog.md

@@ -22,8 +22,9 @@ This catalog names common multi-agent workflows by area and lists, for each role
 
 The catalog is a four-level taxonomy: **Area → Workflow → Stage → Role**. Each area groups related workflows, each workflow lists its stages in order, and each role belongs to one stage, with a short domain description and up to five ranked candidate models.
 
-- **Candidates are dated research.** The lists come from a documentation snapshot taken on 2026-09-06. Prices (USD per million input and output tokens), context windows and capability notes were not re-verified after that date, except for the retired writer model below. Verify them before you dispatch work, and read every quality note as a research claim, not a guarantee.
-- **Writer #2 is Qwen3.7 Plus.** `qwen/qwen3-coder-plus` left the OpenRouter catalog (404) and is not a roster route. Candidate lines that named it now name `qwen/qwen3.7-plus` at the 2026-10-17 list price, $0.32 input and $1.28 output per 1M tokens, 1M context. This repository's writer #2 is the route `pi-qwen3-7-plus-openrouter` (model `openrouter/qwen/qwen3.7-plus`). The writer roster, in order, is `grok-grok-4-7`, `pi-qwen3-7-plus-openrouter`, `agy-gemini-4-argon-high`, `agy-gemini-4-argon-medium`, `agy-gemini-3-8-flash-medium`, `agy-gemini-3-8-flash-high`. [Harness routing](harness-routing.md) is the admission record.
+- **Candidates are dated research.** Prices (USD per million input and output tokens) and context windows were re-verified against the OpenRouter models list (`https://openrouter.ai/api/v1/models`) on 2026-10-29. Capability notes stay research claims, not guarantees. Verify them again before you dispatch work.
+- **Writer #2 is Qwen3.7 Plus.** `qwen/qwen3-coder-plus` left the OpenRouter catalog and is not a roster route. Candidate lines that named it name `qwen/qwen3.7-plus`. Re-verified 2026-10-29: $0.32 input and $1.28 output per 1M tokens, 1M context. This repository's writer #2 is the route `pi-qwen3-7-plus-openrouter` (model `openrouter/qwen/qwen3.7-plus`). The writer roster, in order, is `grok-grok-4-7`, `pi-qwen3-7-plus-openrouter`, `agy-gemini-4-argon-high`, `agy-gemini-4-argon-medium`, `agy-gemini-3-8-flash-medium`, `agy-gemini-3-8-flash-high`. [Harness routing](harness-routing.md) is the admission record.
+- **Removed OpenRouter ids were replaced on 2026-10-29.** `openai/o3-mini-high` is gone. The semantic equivalence verifier now ranks `openai/gpt-6.1-sol` first, the CLI critic in [harness routing](harness-routing.md), and keeps `openai/gpt-5.6-sol` next. `kwaipilot/kat-coder-pro-v2.5` is gone; those slots name `qwen/qwen3-coder`. Catalog roles that named Grok 4.6 now name `x-ai/grok-4.7`. The writer roster is unchanged.
 - **Candidates are not admission.** A model ID here is a research identifier, not a route selector. Your project admits routes in `.kxm/routes.yaml`, and `kxm harness list` shows which harnesses are installed and signed in. [Harness routing](harness-routing.md) explains when to use a native harness and when an aggregator.
 - **Selection is per role.** Areas are navigation only; they never pool unrelated roles into one ranking. Model, harness, platform, modality and required tools are routing attributes, not areas.
 - **Gates stay deterministic.** Where a role is a critic next to a gate, the gate (tests, type checks, linters) decides pass or fail, and the model reviews what a gate cannot check.
@@ -34,9 +35,9 @@ An interactive `kxm init` offers to install the Software Engineering workflows w
 
 | Tier | Price band | Context window | Example models | Typical work |
 |---|---|---|---|---|
-| **Tier 0: Filter and 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 0: Filter and ingestion** | $0.03 - $0.20 / M | 1.0M - 1.05M | 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** | $0.30 - $1.00 / M | 262k - 1.05M | Gemini-3.8-Flash, Qwen-3.7-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 and 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 2: Precision and critic** | $1.25 - $4.00 / M | 200k - 1.05M | GPT-5.6-Sol, Grok-4.7, Claude Sonnet 5, OpenAI o3, DeepSeek-R1, Codex 5.3 | Primary code writing, formal contract generation, concurrency race diagnosis, and causal inference. |
 | **Tier 3: Architecture** | $5.00 - $15.00 / M | 1.0M - 1.05M | Claude Fable 5.1, Claude Opus 5, GPT-6-Astra-Pro (escalation only) | Architecture review, executive briefs, high-stakes legal redlining, and RFC review. |
 
 ---
@@ -131,9 +132,9 @@ This cross-reference points each software and security workflow at the KXM pages
 *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) — Systems planning; 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.
+2. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — 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.
+4. **`x-ai/grok-4.7`** (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
@@ -142,9 +143,9 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`openai/gpt-5.6-sol`** (1.05M ctx | $4.00 / $20.00) — Premier schema contract generator; 100% compliant OpenAPI 3.1, Protobuf v3, and JSON schemas.
+2. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — High-throughput Zod/Pydantic validation pipeline synthesis.
+3. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — Fast translation of data dictionaries into TypeScript interfaces.
 4. **`qwen/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
 
@@ -154,11 +155,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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) — Fast repository builder; resolves path aliases, links, and workspace configs cleanly.
+1. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — Fast repository builder; resolves path aliases, links, 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/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
+5. **`qwen/qwen3-coder`** (262k ctx | $0.30 / $1.00) — Resilient project scaffolding and bundler configuration.
 
 #### Role: Backend Data Implementer
 
@@ -166,11 +167,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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) — Strong first-pass implementation that passes deterministic verification gates with minimal rework.
+1. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — Strong first-pass implementation that passes deterministic verification gates with minimal rework.
 2. **`qwen/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
+3. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — 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.
+5. **`qwen/qwen3-coder`** (262k ctx | $0.30 / $1.00) — Multi-file backend logic implementation adhering to project coding patterns.
 
 #### Role: Frontend Fullstack Implementer
 
@@ -179,8 +180,8 @@ This cross-reference points each software and security workflow at the KXM pages
 *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.
+2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — Fast streaming UI prototyping, token parsing, and JSX/TSX tree generation.
+3. **`x-ai/grok-4.7`** (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.
 
@@ -190,11 +191,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`openai/gpt-5.6-sol`** (1.05M ctx | $4.00 / $20.00) — Flawless SDK synthesis, webhook signature validation, and API adapter error handling.
+2. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — Robust glue code, rate-limiting backoff algorithms, and token refresh loops.
 3. **`qwen/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
+4. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — Low-cost mapping of disparate vendor payloads into internal domain models.
+5. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — High-speed webhook listener and protocol transformation adapter.
 
 ---
 
@@ -211,10 +212,10 @@ This cross-reference points each software and security workflow at the KXM pages
 *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.
+2. **`z-ai/glm-5.3`** (1.05M ctx | $0.0699 / $1.40) — 1.05M 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.
+4. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — Cost-efficient full-codebase impact analysis.
+5. **`moonshotai/kimi-k3`** (1.05M ctx | $0.239 / $10.80) — Multi-package refactoring plan generation.
 
 #### Role: Code Transform Implementer
 
@@ -222,7 +223,7 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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) — Fast execution of multi-file semantic code changes.
+1. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — 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.
@@ -234,9 +235,9 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *Domain:* Critic-with-gate: reviews public contract compliance and behavioral equivalence; deterministic checks (type checks, tests, build output) remain the fixed gate and are never delegated to a model.
 
-1. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — Reviews compiler 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.
+1. **`openai/gpt-6.1-sol`** (1.05M ctx | $2.00 / $10.00) — Deep symbolic verification that AST refactors maintain behavioral equivalence. This is the CLI critic tier in harness routing.
+2. **`openai/gpt-5.6-sol`** (1.05M ctx | $4.00 / $20.00) — Reviews compiler and type-checker cascades against API contracts.
+3. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — 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.
 
@@ -256,7 +257,7 @@ This cross-reference points each software and security workflow at the KXM pages
 
 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.
+3. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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.
 
@@ -278,7 +279,7 @@ This cross-reference points each software and security workflow at the KXM pages
 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.
+5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — Cost-effective evaluation of microservice communication topologies.
 
 #### Role: Scalability Database Topologist
 
@@ -286,11 +287,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — 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.
+5. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — Pragmatic cache tiering (Redis/Dragonfly) and read-replica strategies.
 
 #### Role: Security Architect Threat Modeler
 
@@ -299,10 +300,10 @@ This cross-reference points each software and security workflow at the KXM pages
 *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) — Architecture and permissions review; zero-trust network modeling.
-2. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — Auth token and protocol threat analysis.
+2. **`openai/gpt-5.6-sol`** (1.05M ctx | $4.00 / $20.00) — 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.
+5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — Full-infrastructure vulnerability surface analysis.
 
 #### Role: IaC Cloud Topology Engineer
 
@@ -310,10 +311,10 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — Fast, deterministic Terraform/Kubernetes manifest implementation.
 2. **`qwen/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
+4. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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
@@ -324,9 +325,9 @@ This cross-reference points each software and security workflow at the KXM pages
 
 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.
+3. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — 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.
+5. **`moonshotai/kimi-k3`** (1.05M ctx | $0.239 / $10.80) — Cross-datacenter failover verification.
 
 #### Role: Architecture Critic Protocol Auditor
 
@@ -335,8 +336,8 @@ This cross-reference points each software and security workflow at the KXM pages
 *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) — Architecture and permissions criticism.
-2. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — CLI, protocol and contract criticism.
-3. **`z-ai/glm-5.3`** (1.31M ctx | $1.40 / $4.40) — 1.31M context full-stack protocol audit.
+2. **`openai/gpt-5.6-sol`** (1.05M ctx | $4.00 / $20.00) — CLI, protocol and contract criticism.
+3. **`z-ai/glm-5.3`** (1.05M ctx | $0.0699 / $1.40) — 1.05M 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.
 
@@ -355,10 +356,10 @@ This cross-reference points each software and security workflow at the KXM pages
 *Domain:* Read-only audit of developer and operator documentation against CLI `--help` output, source environment variables, tool schemas, and repository configuration before writing begins.
 
 1. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Audits authority, permissions, and lifecycle models against actual behavior.
-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.
+2. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — High-throughput cross-referencing between source code symbols and markdown documentation.
+3. **`openai/gpt-5.6-sol`** (1.05M ctx | $4.00 / $20.00) — 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.
+5. **`z-ai/glm-5.3`** (1.05M ctx | $0.0699 / $1.40) — Long-context consistency verification across multi-document doc trees.
 
 #### Role: Documentation Writer
 
@@ -366,11 +367,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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) — Rapid, structured Markdown authoring that follows repository conventions.
+1. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — Rapid, structured Markdown authoring that follows repository conventions.
 2. **`qwen/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
+5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — Cost-effective multi-page documentation generation and reference expansion.
 
 #### Role: Documentation Reviewer
 
@@ -379,9 +380,9 @@ This cross-reference points each software and security workflow at the KXM pages
 *Domain:* Critic-with-gate: the architecture critic verifies authority and permission wording; the CLI critic verifies command and flag accuracy; a deterministic docs lint remains the fixed gate.
 
 1. **`anthropic/claude-fable-5.1`** (1M ctx | $10.00 / $50.00) — Verifies security, governance, and authority wording.
-2. **`openai/gpt-5.6-sol`** (1.05M ctx | $2.00 / $10.00) — Verifies CLI command accuracy, flag syntax, and documentation formatting.
+2. **`openai/gpt-5.6-sol`** (1.05M ctx | $4.00 / $20.00) — 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.
+4. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — 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.
 
 ---
@@ -414,7 +415,7 @@ This cross-reference points each software and security workflow at the KXM pages
 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/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
+4. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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
@@ -425,9 +426,9 @@ This cross-reference points each software and security workflow at the KXM pages
 
 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.
+3. **`x-ai/grok-4.7`** (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.
+5. **`deepseek/deepseek-v4-pro`** (1.05M ctx | $1.60 / $3.20) — Bulk UI component library scaffolding.
 
 ---
 
@@ -445,9 +446,9 @@ This cross-reference points each software and security workflow at the KXM pages
 
 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.
+3. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — Fast mobile component implementation and state binding.
+4. **`moonshotai/kimi-k2.7-code`** (262k ctx | $0.56 / $3.10) — Resilient mobile code generation.
+5. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — Mobile screen layout translation and asset mapping.
 
 #### Role: Native Haptics Specialist
 
@@ -475,7 +476,7 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`x-ai/grok-4.7`** (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) — Low-cost candidate; 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.
@@ -488,7 +489,7 @@ This cross-reference points each software and security workflow at the KXM pages
 *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.
+2. **`openai/gpt-5.6-sol`** (1.05M ctx | $4.00 / $20.00) — Flagship CLI and sub-process navigation specialist; signal cleanup.
 3. **`qwen/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
@@ -507,7 +508,7 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — **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.
@@ -520,7 +521,7 @@ This cross-reference points each software and security workflow at the KXM pages
 *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.
+2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — Automated WCAG 2.2 color contrast and visual touch-target auditor.
 3. **`qwen/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
@@ -553,7 +554,7 @@ This cross-reference points each software and security workflow at the KXM pages
 
 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.
+3. **`x-ai/grok-4.7`** (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.
 
@@ -563,11 +564,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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.
+5. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $4.00 / $20.00) — Strict schema compliance for programmatic video generation APIs.
 
 #### Role: Visual Continuity Engine
 
@@ -576,9 +577,9 @@ This cross-reference points each software and security workflow at the KXM pages
 *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.
+2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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.
+4. **`z-ai/glm-5.3-flash`** (1.05M ctx | $0.15 / $0.50) — 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
@@ -587,8 +588,8 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`z-ai/glm-5.3-flash`** (1.05M ctx | $0.15 / $0.50) — **Tier-0 Primary Ingestion Gate**; 1.05M native video context at $0.15/M prompt. The OpenRouter list price for this id is volatile.
+2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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.
@@ -599,11 +600,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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.
+5. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $4.00 / $20.00) — Orchestrates multi-channel frequency and ducking parameter hierarchies.
 
 ---
 
@@ -619,11 +620,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — Scans multi-hour video/audio streams; extracts timestamped clips at sub-$0.50 cost per hour.
+2. **`z-ai/glm-5.3-flash`** (1.05M ctx | $0.15 / $0.50) — 1.05M 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.
+5. **`moonshotai/kimi-k3`** (1.05M ctx | $0.239 / $10.80) — Long-horizon needle-in-a-haystack topic retrieval across long discussions.
 
 #### Role: Reframing B-Roll Matcher
 
@@ -631,9 +632,9 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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.
+3. **`xiaomi/mimo-v2.5`** (1.05M ctx | $0.40 / $2.00) — 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.
 
@@ -643,11 +644,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`x-ai/grok-4.7`** (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.
+5. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $4.00 / $20.00) — Precise timing verification and deterministic subtitle schema alignment.
 
 ---
 
@@ -666,7 +667,7 @@ This cross-reference points each software and security workflow at the KXM pages
 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/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
+4. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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
@@ -676,10 +677,10 @@ This cross-reference points each software and security workflow at the KXM pages
 *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.
+2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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.
+5. **`z-ai/glm-5.3-flash`** (1.05M ctx | $0.15 / $0.50) — High-throughput preliminary broadcast screening.
 
 ---
 
@@ -709,9 +710,9 @@ This cross-reference points each software and security workflow at the KXM pages
 
 1. **`openai/gpt-5.3-codex`** (400k ctx | $1.75 / $14.00) — Complex nested SQL queries, CTEs, and window functions.
 2. **`qwen/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
+3. **`deepseek/deepseek-v4-pro`** (1.05M ctx | $1.60 / $3.20) — 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.
+5. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — Ultra-fast schema resolution and join mapping.
 
 #### Role: SQL Optimizer Validator
 
@@ -731,11 +732,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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.
+3. **`z-ai/glm-5.3-flash`** (1.05M ctx | $0.15 / $0.50) — 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.
+5. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $4.00 / $20.00) — Strict schema compliance for BI embedded frameworks.
 
 ---
 
@@ -755,7 +756,7 @@ This cross-reference points each software and security workflow at the KXM pages
 2. **`qwen/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
+5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — Large-scale data ingestion and transformation scripts.
 
 #### Role: Statistical Causal Reasoner
 
@@ -783,11 +784,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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.
+5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — Low-cost long-form structured financial data normalization.
 
 #### Role: Financial Variance Auditor
 
@@ -798,7 +799,7 @@ This cross-reference points each software and security workflow at the KXM pages
 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.
+4. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $4.00 / $20.00) — 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.
 
 ---
@@ -829,7 +830,7 @@ This cross-reference points each software and security workflow at the KXM pages
 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.
+4. **`x-ai/grok-4.7`** (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
@@ -854,7 +855,7 @@ This cross-reference points each software and security workflow at the KXM pages
 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.
+5. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — Continuous automated monitoring of competitor site changes.
 
 ---
 
@@ -873,7 +874,7 @@ This cross-reference points each software and security workflow at the KXM pages
 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.
+4. **`openai/gpt-5.6-sol`** (1.05M ctx | $4.00 / $20.00) — Term sheet covenant structuring.
 5. **`openai/gpt-6-astra`** (1.05M ctx | $10.00 / $50.00) — Complex multi-party merger and joint venture dynamics.
 
 ---
@@ -902,8 +903,8 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — Sub-second deterministic tool dispatch with zero argument hallucination.
+2. **`openai/gpt-5.6-sol`** (1.05M ctx | $4.00 / $20.00) — 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.
@@ -914,10 +915,10 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`deepseek/deepseek-v4-flash-0731`** (1.05M ctx | $0.119 / $0.238) — High-volume Tier-0 spam/priority email triage at $0.119/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.
+4. **`mistralai/mistral-small-3.2-24b-instruct`** (256k 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
@@ -927,10 +928,10 @@ This cross-reference points each software and security workflow at the KXM pages
 *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.
+2. **`upstage/solar-pro4`** (524k ctx | $0.15 / $0.60) — 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.
+5. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — Multimodal briefing compiling audio voicemails and visual slide decks.
 
 ---
 
@@ -950,7 +951,7 @@ This cross-reference points each software and security workflow at the KXM pages
 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.
+5. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — Direct, transparent post-mortem communication.
 
 ---
 
@@ -998,8 +999,8 @@ This cross-reference points each software and security workflow at the KXM pages
 *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.
+2. **`deepseek/deepseek-v4-flash-0731`** (1.05M ctx | $0.119 / $0.238) — High-throughput anomaly classification in 1.05M context.
+3. **`z-ai/glm-5.3-flash`** (1.05M ctx | $0.15 / $0.50) — 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.
 
@@ -1009,8 +1010,8 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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).
+1. **`deepseek/deepseek-v4-flash-0731`** (1.05M ctx | $0.119 / $0.238) — **Tier-0 Primary Ingestion Gate**; 1.05M context at $0.119/M prompt.
+2. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — 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.
@@ -1033,7 +1034,7 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — Fast, repo-accurate test writer; minimal boilerplate without hallucinating dependencies.
 2. **`qwen/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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).
@@ -1047,9 +1048,9 @@ This cross-reference points each software and security workflow at the KXM pages
 
 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.
+3. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — Root cause extraction without corporate sugarcoating.
+4. **`google/gemini-3.8-flash`** (1.05M ctx | $0.825 / $4.125) — Visual timeline generation combining metric spikes and log traces.
+5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — Infrastructure remediations and preventive patch specification.
 
 ---
 
@@ -1066,7 +1067,7 @@ This cross-reference points each software and security workflow at the KXM pages
 *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 deterministic verification cleanly.
+2. **`x-ai/grok-4.7`** (500k ctx | $2.00 / $6.00) — Rapid patch application that passes deterministic verification cleanly.
 3. **`qwen/qwen3.7-plus`** (1M ctx | $0.32 / $1.28) — 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.
@@ -1077,11 +1078,11 @@ This cross-reference points each software and security workflow at the KXM pages
 
 *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.
+1. **`openai/gpt-5.6-sol-pro`** (1.05M ctx | $4.00 / $20.00) — 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.
+4. **`nvidia/nemotron-3-ultra-550b-a55b`** (262k ctx | $0.60 / $2.40) — 550B MoE frontier open reasoning model for vulnerability search.
+5. **`deepseek/deepseek-v4-pro-0813`** (1.05M ctx | $0.66 / $1.98) — Full-codebase taint tracking and dataflow analysis.
 
 ---
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@kontextmind/kxm",
-  "version": "0.7.208",
+  "version": "0.7.210",
   "description": "KXM local-first multi-agent orchestration and operator dashboard",
   "type": "module",
   "author": "KontextMind",

package/plugins/kxm/.claude-plugin/plugin.json

@@ -2,7 +2,7 @@
   "$schema": "https://json.schemastore.org/claude-code-plugin-manifest.json",
   "name": "kxm",
   "displayName": "KXM",
-  "version": "0.7.208",
+  "version": "0.7.210",
   "description": "Headless multi-agent orchestration, durable workflows, and a live operator dashboard for Pi and Claude Code",
   "author": {
     "name": "KontextMind",

package/plugins/kxm/dist/cli.js

@@ -57068,9 +57068,10 @@ import { dirname as dirname25, join as join58 } from "node:path";
 var ADMITTED_GUIDE_BINDINGS = Object.freeze({
   "anthropic/claude-fable-5.1": { harness: "claude", provider: "anthropic", model: "fable" },
   "anthropic/fable": { harness: "claude", provider: "anthropic", model: "fable" },
+  "openai/gpt-6.1-sol": { harness: "codex", provider: "openai", model: "gpt-6.1-sol" },
   "openai/gpt-5.6-sol": { harness: "codex", provider: "openai", model: "gpt-5.6-sol" },
-  "x-ai/grok-4.6": { harness: "grok", provider: "xai", model: "grok-4.6" },
-  "xai/grok-4.6": { harness: "grok", provider: "xai", model: "grok-4.6" },
+  "x-ai/grok-4.7": { harness: "grok", provider: "xai", model: "grok-4.7" },
+  "xai/grok-4.7": { harness: "grok", provider: "xai", model: "grok-4.7" },
   "qwen/qwen3.7-plus": { harness: "pi", provider: "openrouter", model: "qwen/qwen3.7-plus" }
 });
 var GUIDE_WORKFLOWS = [
@@ -57089,7 +57090,7 @@ var GUIDE_WORKFLOWS = [
           { vendor: "anthropic", model: "claude-fable-5.1" },
           { vendor: "deepseek", model: "deepseek-v4-pro-0813" },
           { vendor: "openai", model: "gpt-6-astra-pro" },
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "qwen", model: "qwen3.8-max-0902" }
         ]
       },
@@ -57112,11 +57113,11 @@ var GUIDE_WORKFLOWS = [
         role: "scaffold-build-specialist",
         domain: "Provisions monorepo workspace topologies, linters, Docker multi-stage builds, and CI pipelines.",
         candidates: [
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "mistralai", model: "devstral-2512" },
           { vendor: "qwen", model: "qwen3.7-plus" },
           { vendor: "openai", model: "gpt-5.3-codex" },
-          { vendor: "kwaipilot", model: "kat-coder-pro-v2.5" }
+          { vendor: "qwen", model: "qwen3-coder" }
         ]
       },
       {
@@ -57125,11 +57126,11 @@ var GUIDE_WORKFLOWS = [
         role: "backend-data-implementer",
         domain: "Implements domain entities, SQL queries/migrations, ORM persistence, concurrency primitives, and transactional boundaries.",
         candidates: [
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "qwen", model: "qwen3.7-plus" },
           { vendor: "deepseek", model: "deepseek-v4-pro-0813" },
           { vendor: "openai", model: "gpt-5.3-codex" },
-          { vendor: "kwaipilot", model: "kat-coder-pro-v2.5" }
+          { vendor: "qwen", model: "qwen3-coder" }
         ]
       },
       {
@@ -57140,7 +57141,7 @@ var GUIDE_WORKFLOWS = [
         candidates: [
           { vendor: "anthropic", model: "claude-sonnet-5" },
           { vendor: "google", model: "gemini-3.8-flash" },
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "bytedance-seed", model: "seed-2.0-code" },
           { vendor: "openai", model: "gpt-5.3-codex" }
         ]
@@ -57152,7 +57153,7 @@ var GUIDE_WORKFLOWS = [
         domain: "Wires external APIs, authentication flows (OAuth2/OIDC), webhook validation, cloud SDKs, and async worker queues.",
         candidates: [
           { vendor: "openai", model: "gpt-5.6-sol" },
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "qwen", model: "qwen3.7-plus" },
           { vendor: "deepseek", model: "deepseek-v4-pro-0813" },
           { vendor: "google", model: "gemini-3.8-flash" }
@@ -57185,7 +57186,7 @@ var GUIDE_WORKFLOWS = [
         role: "code-transform-implementer",
         domain: "Executes surgical multi-file edits, applying design patterns while preserving comments and formatting.",
         candidates: [
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "relace", model: "relace-apply-3" },
           { vendor: "openai", model: "gpt-5.3-codex" },
           { vendor: "morph", model: "morph-v3-large" },
@@ -57198,8 +57199,8 @@ var GUIDE_WORKFLOWS = [
         role: "semantic-equivalence-verifier",
         domain: "Critic-with-gate: reviews public contract compliance and behavioral equivalence; deterministic checks remain the fixed witness gate and are never delegated to an LLM.",
         candidates: [
+          { vendor: "openai", model: "gpt-6.1-sol" },
           { vendor: "openai", model: "gpt-5.6-sol" },
-          { vendor: "openai", model: "o3-mini-high" },
           { vendor: "deepseek", model: "deepseek-v4-pro-0813" },
           { vendor: "anthropic", model: "claude-opus-5" },
           { vendor: "mistralai", model: "devstral-2512" }
@@ -57257,7 +57258,7 @@ var GUIDE_WORKFLOWS = [
           { vendor: "openai", model: "gpt-6-astra-pro" },
           { vendor: "anthropic", model: "claude-fable-5.1" },
           { vendor: "qwen", model: "qwen3.8-2.4t-a95b" },
-          { vendor: "x-ai", model: "grok-4.6" }
+          { vendor: "x-ai", model: "grok-4.7" }
         ]
       },
       {
@@ -57279,7 +57280,7 @@ var GUIDE_WORKFLOWS = [
         role: "iac-cloud-topology-engineer",
         domain: "Generates declarative Terraform/OpenTofu, Pulumi, Kubernetes CRDs, and VPC networking definitions.",
         candidates: [
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "qwen", model: "qwen3.7-plus" },
           { vendor: "openai", model: "gpt-5.3-codex" },
           { vendor: "google", model: "gemini-3.8-flash" },
@@ -57339,7 +57340,7 @@ var GUIDE_WORKFLOWS = [
         role: "documentation-writer",
         domain: "Authors concise, accurate developer and operator documentation with clear terminology, correct command flags, and consistent formatting without marketing drift.",
         candidates: [
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "qwen", model: "qwen3.7-plus" },
           { vendor: "anthropic", model: "claude-sonnet-5" },
           { vendor: "openai", model: "gpt-5.3-codex" },

package/plugins/kxm/dist/mcp-server.js

@@ -25462,7 +25462,7 @@ function sessionTokenFixHint(policy) {
 }
 
 // plugins/kxm/src/mcp-server.ts
-var VERSION = "0.7.208";
+var VERSION = "0.7.210";
 var CONFIGURE_PLUGIN = "/plugin configure kxm@kxm";
 var inbox = /* @__PURE__ */ new Map();
 var notifiedInbox = /* @__PURE__ */ new Set();

package/plugins/kxm/package.json

@@ -1,6 +1,6 @@
 {
   "name": "claude-plugin",
-  "version": "0.7.208",
+  "version": "0.7.210",
   "private": true,
   "type": "module",
   "engines": {

package/plugins/kxm/src/init-guide-setup.ts

@@ -61,9 +61,10 @@ export interface GuideWorkflow {
 const ADMITTED_GUIDE_BINDINGS: Readonly<Record<string, AgentBinding>> = Object.freeze({
   "anthropic/claude-fable-5.1": { harness: "claude", provider: "anthropic", model: "fable" },
   "anthropic/fable": { harness: "claude", provider: "anthropic", model: "fable" },
+  "openai/gpt-6.1-sol": { harness: "codex", provider: "openai", model: "gpt-6.1-sol" },
   "openai/gpt-5.6-sol": { harness: "codex", provider: "openai", model: "gpt-5.6-sol" },
-  "x-ai/grok-4.6": { harness: "grok", provider: "xai", model: "grok-4.6" },
-  "xai/grok-4.6": { harness: "grok", provider: "xai", model: "grok-4.6" },
+  "x-ai/grok-4.7": { harness: "grok", provider: "xai", model: "grok-4.7" },
+  "xai/grok-4.7": { harness: "grok", provider: "xai", model: "grok-4.7" },
   "qwen/qwen3.7-plus": { harness: "pi", provider: "openrouter", model: "qwen/qwen3.7-plus" },
 });
 
@@ -84,7 +85,7 @@ export const GUIDE_WORKFLOWS: readonly GuideWorkflow[] = [
           { vendor: "anthropic", model: "claude-fable-5.1" },
           { vendor: "deepseek", model: "deepseek-v4-pro-0813" },
           { vendor: "openai", model: "gpt-6-astra-pro" },
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "qwen", model: "qwen3.8-max-0902" },
         ],
       },
@@ -107,11 +108,11 @@ export const GUIDE_WORKFLOWS: readonly GuideWorkflow[] = [
         role: "scaffold-build-specialist",
         domain: "Provisions monorepo workspace topologies, linters, Docker multi-stage builds, and CI pipelines.",
         candidates: [
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "mistralai", model: "devstral-2512" },
           { vendor: "qwen", model: "qwen3.7-plus" },
           { vendor: "openai", model: "gpt-5.3-codex" },
-          { vendor: "kwaipilot", model: "kat-coder-pro-v2.5" },
+          { vendor: "qwen", model: "qwen3-coder" },
         ],
       },
       {
@@ -120,11 +121,11 @@ export const GUIDE_WORKFLOWS: readonly GuideWorkflow[] = [
         role: "backend-data-implementer",
         domain: "Implements domain entities, SQL queries/migrations, ORM persistence, concurrency primitives, and transactional boundaries.",
         candidates: [
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "qwen", model: "qwen3.7-plus" },
           { vendor: "deepseek", model: "deepseek-v4-pro-0813" },
           { vendor: "openai", model: "gpt-5.3-codex" },
-          { vendor: "kwaipilot", model: "kat-coder-pro-v2.5" },
+          { vendor: "qwen", model: "qwen3-coder" },
         ],
       },
       {
@@ -135,7 +136,7 @@ export const GUIDE_WORKFLOWS: readonly GuideWorkflow[] = [
         candidates: [
           { vendor: "anthropic", model: "claude-sonnet-5" },
           { vendor: "google", model: "gemini-3.8-flash" },
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "bytedance-seed", model: "seed-2.0-code" },
           { vendor: "openai", model: "gpt-5.3-codex" },
         ],
@@ -147,7 +148,7 @@ export const GUIDE_WORKFLOWS: readonly GuideWorkflow[] = [
         domain: "Wires external APIs, authentication flows (OAuth2/OIDC), webhook validation, cloud SDKs, and async worker queues.",
         candidates: [
           { vendor: "openai", model: "gpt-5.6-sol" },
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "qwen", model: "qwen3.7-plus" },
           { vendor: "deepseek", model: "deepseek-v4-pro-0813" },
           { vendor: "google", model: "gemini-3.8-flash" },
@@ -180,7 +181,7 @@ export const GUIDE_WORKFLOWS: readonly GuideWorkflow[] = [
         role: "code-transform-implementer",
         domain: "Executes surgical multi-file edits, applying design patterns while preserving comments and formatting.",
         candidates: [
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "relace", model: "relace-apply-3" },
           { vendor: "openai", model: "gpt-5.3-codex" },
           { vendor: "morph", model: "morph-v3-large" },
@@ -193,8 +194,8 @@ export const GUIDE_WORKFLOWS: readonly GuideWorkflow[] = [
         role: "semantic-equivalence-verifier",
         domain: "Critic-with-gate: reviews public contract compliance and behavioral equivalence; deterministic checks remain the fixed witness gate and are never delegated to an LLM.",
         candidates: [
+          { vendor: "openai", model: "gpt-6.1-sol" },
           { vendor: "openai", model: "gpt-5.6-sol" },
-          { vendor: "openai", model: "o3-mini-high" },
           { vendor: "deepseek", model: "deepseek-v4-pro-0813" },
           { vendor: "anthropic", model: "claude-opus-5" },
           { vendor: "mistralai", model: "devstral-2512" },
@@ -252,7 +253,7 @@ export const GUIDE_WORKFLOWS: readonly GuideWorkflow[] = [
           { vendor: "openai", model: "gpt-6-astra-pro" },
           { vendor: "anthropic", model: "claude-fable-5.1" },
           { vendor: "qwen", model: "qwen3.8-2.4t-a95b" },
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
         ],
       },
       {
@@ -274,7 +275,7 @@ export const GUIDE_WORKFLOWS: readonly GuideWorkflow[] = [
         role: "iac-cloud-topology-engineer",
         domain: "Generates declarative Terraform/OpenTofu, Pulumi, Kubernetes CRDs, and VPC networking definitions.",
         candidates: [
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "qwen", model: "qwen3.7-plus" },
           { vendor: "openai", model: "gpt-5.3-codex" },
           { vendor: "google", model: "gemini-3.8-flash" },
@@ -334,7 +335,7 @@ export const GUIDE_WORKFLOWS: readonly GuideWorkflow[] = [
         role: "documentation-writer",
         domain: "Authors concise, accurate developer and operator documentation with clear terminology, correct command flags, and consistent formatting without marketing drift.",
         candidates: [
-          { vendor: "x-ai", model: "grok-4.6" },
+          { vendor: "x-ai", model: "grok-4.7" },
           { vendor: "qwen", model: "qwen3.7-plus" },
           { vendor: "anthropic", model: "claude-sonnet-5" },
           { vendor: "openai", model: "gpt-5.3-codex" },

package/plugins/kxm/src/mcp-server.ts

@@ -11,7 +11,7 @@ import { deliverInboxNotification } from "./inbox.ts";
 import type { HubEvent, MessageRecord } from "./protocol.ts";
 import { sessionTokenFixHint } from "./session-token-hint.ts";
 
-const VERSION = "0.7.208";
+const VERSION = "0.7.210";
 const CONFIGURE_PLUGIN = "/plugin configure kxm@kxm";
 const inbox = new Map<string, MessageRecord>();
 const notifiedInbox = new Set<string>();