@zaip-ai/cli 4.2.0 → 4.4.0

package/templates/.agents/README.md

@@ -879,6 +879,14 @@ O modelo de ameaça, a matriz allow/deny, todas as camadas, limitações, opera
 rollout, rollback e comandos de verificação estão documentados no
 [guia completo de segurança dos perfis ZAIP-AIOS](zaip-aios/guides/10-seguranca.md).
 
+Na qualificação dos hooks, diferencie falha técnica, falso bloqueio e recusa de
+segurança esperada. Exija testes do pacote empacotado e uma edição real por
+subagente nativo com o hook ativo; testes simulados não comprovam a identidade
+que o host entrega. O [guia de diagnóstico](zaip-aios/guides/14-troubleshooting.md)
+descreve os formatos de sessão/turno, o contrato JSON de Stop e a recuperação
+limitada por recibo. Qualificação local não significa publicação no npm nem
+garantia de ausência de bugs em todo ambiente.
+
 
 ## Automação e integrações opcionais
 

package/templates/.agents/config/agents/discovery.json

@@ -1,5 +1,5 @@
 {
-  "_zaip_managed": "zaip-managed generated-do-not-edit source=agent-discovery version=4.2.0 sha256=3def7a2b43c1a751f8e68754ef94cb107a80b5d26e11f065af2b4cc145fc8897",
+  "_zaip_managed": "zaip-managed generated-do-not-edit source=agent-discovery version=4.4.0 sha256=3def7a2b43c1a751f8e68754ef94cb107a80b5d26e11f065af2b4cc145fc8897",
   "schema_version": 1,
   "source": {
     "path": ".agents/config/agents/catalog.yaml",

package/templates/.agents/config/clients/codex/agents/zaip-agent-backend.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-backend"
 description = "Implement and verify backend, API, persistence, and integration boundaries inside declared scopes."
 nickname_candidates = ["will_backend_specialist_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-backend (ZAIP catalog id agent-backend) with role builder.
@@ -12,6 +10,7 @@ Capabilities: backend-delivery, api-delivery, persistence-integration, meta-devt
 Allowed tools: apply-patch, shell-read, test-runner, meta-devtools-mcp.
 Available skills: zaip-skill-change, zaip-skill-backend, zaip-skill-data-modeling, zaip-skill-database-migrations, zaip-skill-auth-tenancy, zaip-skill-api-contracts, zaip-skill-async-jobs, zaip-skill-cache-strategy, zaip-skill-external-integrations, zaip-skill-storage-uploads, zaip-skill-performance-engineering, zaip-skill-testing-strategy, zaip-skill-meta-devtools-api-integration; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: apps/backend/*, modules/*.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: task, requirements, context-pack.

package/templates/.agents/config/clients/codex/agents/zaip-agent-brand-strategist.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-brand-strategist"
 description = "Create, edit, audit, and advise on brand platforms, Brandbooks, identity governance, and unresolved brand decisions inside declared scopes."
 nickname_candidates = ["sarah_branding_specialist_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-brand-strategist (ZAIP catalog id agent-brand-strategist) with role builder.
@@ -12,6 +10,7 @@ Capabilities: brand-strategy, brand-governance, brandbook-authoring, brand-archi
 Allowed tools: apply-patch, shell-read, test-runner.
 Available skills: zaip-skill-change, zaip-skill-brand-governance; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: declared-write-scope.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: request-or-task, requirements, brand-context.

package/templates/.agents/config/clients/codex/agents/zaip-agent-builder.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-builder"
 description = "Deliver generic technical work and SPEC decomposition, including dashboard or local scope when no specialist is mandatory."
 nickname_candidates = ["rafa_tech_lead_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-builder (ZAIP catalog id agent-builder) with role builder.
@@ -12,6 +10,7 @@ Capabilities: project-stack-delivery, test-authoring, spec-creation, spec-refine
 Allowed tools: apply-patch, shell-read, test-runner, zaip-cli.
 Available skills: zaip-skill-change, zaip-skill-spec, zaip-skill-architecture-design, zaip-skill-dev-environment, zaip-skill-module-scaffold, zaip-skill-testing-strategy, zaip-skill-test-data, zaip-skill-dependency-upgrade, zaip-skill-update-version-doc; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: packages/*, .agents/records/specs/*, .agents/records/lifecycle/*, .agents/dashboard/*, .local/*.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: task, requirements, context-pack.

package/templates/.agents/config/clients/codex/agents/zaip-agent-cloud-engineer.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-cloud-engineer"
 description = "Design and validate declared cloud infrastructure, including AWS preparation and future authorized provisioning with scoped AWS CLI and Terraform, and run only contract-bound Cloudflare DNS, Hostinger VPS, SSH, and Terraform operations."
 nickname_candidates = ["dan_cloud_engineer_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-cloud-engineer (ZAIP catalog id agent-cloud-engineer) with role builder.
@@ -12,6 +10,7 @@ Capabilities: cloud-infrastructure, terraform, aws-cli, aws-environment-preparat
 Allowed tools: apply-patch, shell-read, test-runner.
 Available skills: zaip-skill-change, zaip-skill-cloud-engineering, zaip-skill-hostinger-cloudflare-terraform, zaip-skill-container-runtime, zaip-skill-deployment-pipeline, zaip-skill-performance-engineering, zaip-skill-application-observability, zaip-skill-disaster-recovery, zaip-skill-application-readiness; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: declared-write-scope.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: task, requirements, infrastructure-context, autonomous-cloud-contract.

package/templates/.agents/config/clients/codex/agents/zaip-agent-data-analyst.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-data-analyst"
 description = "Analyze bounded datasets, explain methods and uncertainty, and return reproducible evidence-backed findings."
 nickname_candidates = ["gabi_data_analyst_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-data-analyst (ZAIP catalog id agent-data-analyst) with role task-executor.
@@ -12,6 +10,7 @@ Capabilities: data-analysis, metric-diagnostics, evidence-production.
 Allowed tools: shell-read, test-runner.
 Available skills: zaip-skill-change, zaip-skill-data-analyst; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: declared-write-scope.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: task, dataset, metric-definitions.

package/templates/.agents/config/clients/codex/agents/zaip-agent-designer.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-designer"
 description = "Govern the design system for digital products: the central visual rule in DESIGN.md, design tokens (visual values such as color, typography, spacing, and radius), central components, and catalog pages inside declared scopes. Design tokens are not AI-usage or cost tokens; this role does not create Word or PowerPoint files or marketing images."
 nickname_candidates = ["manu_product_designer_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-designer (ZAIP catalog id agent-designer) with role builder.
@@ -12,6 +10,7 @@ Capabilities: design-system, design-tokens, central-components, page-design.
 Allowed tools: apply-patch, shell-read, test-runner.
 Available skills: zaip-skill-change, zaip-skill-design-system, zaip-skill-frontend-componentization; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: .agents/design/*, .agents/design/**.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: task, requirements, design-context.

package/templates/.agents/config/clients/codex/agents/zaip-agent-frontend.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-frontend"
 description = "Implement frontend experiences, accessibility, and component integration using the central visual contract."
 nickname_candidates = ["doug_frontend_specialist_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-frontend (ZAIP catalog id agent-frontend) with role builder.
@@ -12,6 +10,7 @@ Capabilities: frontend-delivery, accessibility, component-integration, meta-devt
 Allowed tools: apply-patch, shell-read, test-runner, meta-devtools-mcp.
 Available skills: zaip-skill-change, zaip-skill-frontend, zaip-skill-frontend-componentization, zaip-skill-frontend-data-flows, zaip-skill-testing-strategy, zaip-skill-meta-devtools-api-integration; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: apps/frontend/*, .agents/dashboard/apps/frontend/*, .agents/dashboard/modules/*/src/react/*.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: task, requirements, .agents/design/DESIGN.md.

package/templates/.agents/config/clients/codex/agents/zaip-agent-general-operations.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-general-operations"
 description = "Deliver bounded general-operations work outside mandatory specialist scopes, including documents, presentations, spreadsheets, process artifacts, and evidence-grounded comparative research."
 nickname_candidates = ["clara_operations_generalist_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-general-operations (ZAIP catalog id agent-general-operations) with role builder.
@@ -12,6 +10,7 @@ Capabilities: general-operations, task-delivery, document-delivery, presentation
 Allowed tools: apply-patch, shell-read, test-runner.
 Available skills: zaip-skill-change, zaip-skill-general-operations; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: declared-write-scope.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: request-or-task, requirements, operations-context.

package/templates/.agents/config/clients/codex/agents/zaip-agent-governance-compliance.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-governance-compliance"
 description = "Prepare operational governance and compliance artifacts for HR, legal operations, policies, privacy, vendors, and risks inside declared scopes."
 nickname_candidates = ["renata_governance_compliance_specialist_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-governance-compliance (ZAIP catalog id agent-governance-compliance) with role builder.
@@ -12,6 +10,7 @@ Capabilities: governance-compliance, task-delivery, hr-artifacts, legal-operatio
 Allowed tools: apply-patch, shell-read, test-runner.
 Available skills: zaip-skill-change, zaip-skill-decisions, zaip-skill-governance-compliance; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: declared-write-scope.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: task, requirements, governance-context.

package/templates/.agents/config/clients/codex/agents/zaip-agent-growth-strategist.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-growth-strategist"
 description = "Advise on business strategy, Growth, RevOps, portfolio priorities, business models, and revenue-or-cost decisions inside declared scopes."
 nickname_candidates = ["robert_growth_strategist_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-growth-strategist (ZAIP catalog id agent-growth-strategist) with role builder.
@@ -12,6 +10,7 @@ Capabilities: growth-strategy, revops-strategy, experimentation-strategy, revenu
 Allowed tools: apply-patch, shell-read, test-runner.
 Available skills: zaip-skill-change, zaip-skill-growth-strategy-revops; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: declared-write-scope.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: request-or-task, requirements, growth-context, business-metrics.

package/templates/.agents/config/clients/codex/agents/zaip-agent-marketing-analyst.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-marketing-analyst"
 description = "Create evidence-grounded marketing strategy, copy, campaigns, content, and assets inside declared scopes."
 nickname_candidates = ["lucia_marketing_analyst_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-marketing-analyst (ZAIP catalog id agent-marketing-analyst) with role builder.
@@ -12,6 +10,7 @@ Capabilities: marketing-strategy, copywriting, campaign-planning, marketing-asse
 Allowed tools: apply-patch, shell-read, imagegen, meta-ads-mcp.
 Available skills: zaip-skill-change, zaip-skill-marketing, zaip-skill-meta-ads; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: declared-write-scope.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: request-or-task, requirements, marketing-context, approved-brandbook.

package/templates/.agents/config/clients/codex/agents/zaip-agent-orchestrator.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-orchestrator"
 description = "Resolve the DAG and delegate only to capable least-privilege agents."
 nickname_candidates = ["dave_product_owner_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "read-only"
 developer_instructions = """
 Operate as Codex agent zaip-agent-orchestrator (ZAIP catalog id agent-orchestrator) with role orchestrator.
@@ -12,6 +10,7 @@ Capabilities: dag-orchestration, lifecycle-governance.
 Allowed tools: zaip-cli, shell-read, tool-discovery, native-collaboration.
 Available skills: zaip-skill-execution; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: read-only.
 This role is read-only. Do not create, modify, move, or delete workspace files.
 Inputs: prepared-plan, active-work.

package/templates/.agents/config/clients/codex/agents/zaip-agent-refactoring-specialist.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-refactoring-specialist"
 description = "Plan and execute safe refactors, migrations, modularization, and architecture improvements from an inventory-backed map with reversible slices and explicit validation gates."
 nickname_candidates = ["tati_refactoring_specialist_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-refactoring-specialist (ZAIP catalog id agent-refactoring-specialist) with role builder.
@@ -12,6 +10,7 @@ Capabilities: refactor-existing-code, migration, modularization, architecture, s
 Allowed tools: apply-patch, shell-read, test-runner.
 Available skills: zaip-skill-change, zaip-skill-refactor-existing-code, zaip-skill-enforce-architecture; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: declared-write-scope.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: task, source-root, target-architecture, baseline.

package/templates/.agents/config/clients/codex/agents/zaip-agent-sales-specialist.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-sales-specialist"
 description = "Create, edit, and advise on internal sales methodology, processes, sales-facing scripts, qualification, enablement, and handoffs inside declared scopes."
 nickname_candidates = ["aaron_sales_specialist_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-sales-specialist (ZAIP catalog id agent-sales-specialist) with role builder.
@@ -12,6 +10,7 @@ Capabilities: sales-strategy, sales-methodology, sales-process, sales-playbook,
 Allowed tools: apply-patch, shell-read, test-runner.
 Available skills: zaip-skill-change, zaip-skill-sales-methodology; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: declared-write-scope.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: request-or-task, requirements, sales-context.

package/templates/.agents/config/clients/codex/agents/zaip-agent-task-executor.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-task-executor"
 description = "Execute one bounded generic Task, including dashboard or local scope, and return verification evidence to the orchestrator."
 nickname_candidates = ["leo_software_engineer_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-task-executor (ZAIP catalog id agent-task-executor) with role task-executor.
@@ -12,6 +10,7 @@ Capabilities: task-delivery, verification, evidence-production.
 Allowed tools: apply-patch, shell-read, test-runner.
 Available skills: zaip-skill-change, zaip-skill-architecture-design, zaip-skill-dev-environment, zaip-skill-module-scaffold, zaip-skill-testing-strategy, zaip-skill-test-data, zaip-skill-systematic-debugging; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: declared-write-scope.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: task, requirements, context-pack.

package/templates/.agents/config/clients/codex/agents/zaip-agent-website-developer.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-website-developer"
 description = "Build website experiences with design-system, SEO, GEO, Markdown for Agents, and Agent Readiness awareness."
 nickname_candidates = ["jonny_website_developer_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-website-developer (ZAIP catalog id agent-website-developer) with role builder.
@@ -12,6 +10,7 @@ Capabilities: website-delivery, design-system, seo, geo, markdown-for-agents, ag
 Allowed tools: apply-patch, shell-read, test-runner, meta-devtools-mcp.
 Available skills: zaip-skill-change, zaip-skill-website-readiness, zaip-skill-frontend-componentization, zaip-skill-meta-devtools-api-integration; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: apps/site/*.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: task, requirements, website-context.

package/templates/.agents/config/clients/codex/agents/zaip-agent-writer.toml

@@ -2,8 +2,6 @@
 name = "zaip-agent-writer"
 description = "Produce faithful summaries, timestamp-aware transcript analyses, and source-grounded documentation for a declared audience without unsupported claims."
 nickname_candidates = ["lia_technical_writer_zaip_agent"]
-model = "gpt-6-sol"
-model_reasoning_effort = "medium"
 sandbox_mode = "workspace-write"
 developer_instructions = """
 Operate as Codex agent zaip-agent-writer (ZAIP catalog id agent-text-summarizer) with role task-executor.
@@ -12,6 +10,7 @@ Capabilities: faithful-summarization, timestamped-transcript-analysis, source-gr
 Allowed tools: apply-patch, shell-read, test-runner.
 Available skills: zaip-skill-context, zaip-skill-change, zaip-skill-text-encoding; load matching skills only. Change/execution skills apply to persistent delivery; Meta Ads applies only to Meta work.
 Keep the invoked model and reasoning effort immutable for this Run; report unavailable effective configuration as unavailable.
+Model and reasoning effort are controlled by the authorized dispatch policy. Catalog preferences are dispatch defaults, not permanent native locks. Before dispatch, the authorized host must select and validate the model/effort allowed by the frozen Task policy and the host's manual selection. Record the requested configuration separately from the effective model/effort observed from the host; never silently inherit a parent configuration or claim a model switch from instructions. If the host cannot select or verify the required configuration, report the capability as unavailable and return the blocked dispatch to the orchestrator.
 Write scopes: declared-write-scope.
 The Codex host grants workspace-write at the workspace boundary, not per-path. Treat the declared write scopes as mandatory: do not modify files outside them; report any required expansion to the parent before writing.
 Inputs: task, source-material, audience, output-constraints.

package/templates/.agents/dashboard/.zaip-managed.json

@@ -1,6 +1,6 @@
 {
   "schemaVersion": 1,
-  "version": "4.2.0",
+  "version": "4.4.0",
   "files": [
     {
       "path": "DATA-DICTIONARY.md",
@@ -15,28 +15,28 @@
       "sha256": "a5ee8934997171e871200e5ab6db33dcef0446c37c9547a3d22f6bb3aff491b1"
     },
     {
-      "path": "apps/frontend/dist/assets/DashboardHelpDialog-DRI_XwUf.css",
-      "sha256": "95dbb5c26ad73831f36f2a51d5e08ef5eee7ca85190764a3e63bf1e700fbadf8"
+      "path": "apps/frontend/dist/assets/DashboardHelpDialog-CkQ_y8kg.js",
+      "sha256": "439758f84a3522a55eb38ba14db5a023d6b135d10ad9671cdd43e46e692708e2"
     },
     {
-      "path": "apps/frontend/dist/assets/DashboardHelpDialog-DlN5EMmD.js",
-      "sha256": "b2f283db1ec9e7e0e3c54f2efb19922bdb3d1df6ec40a452748b0a2f63bdb2c4"
+      "path": "apps/frontend/dist/assets/DashboardHelpDialog-DRI_XwUf.css",
+      "sha256": "95dbb5c26ad73831f36f2a51d5e08ef5eee7ca85190764a3e63bf1e700fbadf8"
     },
     {
       "path": "apps/frontend/dist/assets/PricingSettingsDialog-v9zGRTlk.js",
       "sha256": "b0ff1ac14837810711047d1be79a55e7f91988a5930661094ea79c0f5371d143"
     },
-    {
-      "path": "apps/frontend/dist/assets/ProjectSettingsPage-BRdoeTWk.js",
-      "sha256": "f3706bbe409eed0b35164d931e4f067c5bfb6f3552217ca787d73eea71d02004"
-    },
     {
       "path": "apps/frontend/dist/assets/ProjectSettingsPage-CQMhvzvJ.css",
       "sha256": "10e875c4e3ab6c252983822e25a85db3d707d903c0fcf39812909fdd1d013f48"
     },
     {
-      "path": "apps/frontend/dist/assets/TraceabilityWorkspace-B_5u4Xnp.js",
-      "sha256": "969e137acd41925cf06229bd97e483f569260665e3385e439731443530e25fdd"
+      "path": "apps/frontend/dist/assets/ProjectSettingsPage-doIMhei7.js",
+      "sha256": "28b6167a8e8b74c7aaa09ff600eb31f10e89af0dd8510c79dca40633aa38f63f"
+    },
+    {
+      "path": "apps/frontend/dist/assets/TraceabilityWorkspace-C_4ftjyB.js",
+      "sha256": "baa839e93efa3beeca6904e997d35a04d471287f8b58a5d69d4a2092575508ef"
     },
     {
       "path": "apps/frontend/dist/assets/TraceabilityWorkspace-DvvavDLY.css",
@@ -58,14 +58,14 @@
       "path": "apps/frontend/dist/assets/icons-vendor-Cv3WBFIF.js",
       "sha256": "f2bde355996235413fec4a9fc854f4669f9a233421e104cead02fa2bce7003de"
     },
-    {
-      "path": "apps/frontend/dist/assets/index-C4rLXrz_.js",
-      "sha256": "97e8deebdc4636afcea722941ae974031bca75d3b36ed8fc844541a548636954"
-    },
     {
       "path": "apps/frontend/dist/assets/index-OBJXZOwe.css",
       "sha256": "43de3012ecb2f5ffd2e8096c0543f35b390066d0d2033440b81682b01e560e89"
     },
+    {
+      "path": "apps/frontend/dist/assets/index-Rbmpz4UO.js",
+      "sha256": "c12cea4a30399bff81a6aac1f12e6b935377551f2bc54d508d6c0c44eb404d80"
+    },
     {
       "path": "apps/frontend/dist/assets/logo_novo_azul-BBx5pA2B.png",
       "sha256": "bb6043bda9fc53a111aaea4d6a004559b9ce8a3569691c060d217a7c6aafa0ae"
@@ -92,7 +92,7 @@
     },
     {
       "path": "apps/frontend/dist/index.html",
-      "sha256": "3b8cd7772ee8589aec743d3607c24808feefd1d0cc6e145c14793396d3b67d50"
+      "sha256": "cf07311e4835a7df3e1e0ed767755d7954578204f999381fe33e38e2085e1816"
     },
     {
       "path": "apps/frontend/dist/logo_novo_azul.png",
@@ -104,7 +104,7 @@
     },
     {
       "path": "apps/frontend/package.json",
-      "sha256": "65231c216c31aeade0c984b814397cc999859fafafa0db6d91524518c2f57f09"
+      "sha256": "a84c8070c81247bc04653c9636eb5bc3d66ec64811db08b7a966af5c40de9539"
     },
     {
       "path": "apps/frontend/playwright.dom.config.ts",
@@ -564,7 +564,7 @@
     },
     {
       "path": "modules/cost-monitor/package.json",
-      "sha256": "343ed9bbab867ada90f732bf0bc520c831c4991adb4d209ecb02ec84207e0f19"
+      "sha256": "ba7a473766ca8ba14728e9dd0d3234456de98e49c275d741a3508fa818afbe64"
     },
     {
       "path": "modules/cost-monitor/src/agent-groups.test.ts",
@@ -972,7 +972,7 @@
     },
     {
       "path": "modules/operations/package.json",
-      "sha256": "ed9ab59b13875adabf69fef75f5e4c137e31fe9b94e3e83b2c50b46b78135181"
+      "sha256": "53d48077aaa19ff324d096f27ed8e8f697a409118976667a5012e4f3f3a20cbd"
     },
     {
       "path": "modules/operations/src/contracts.ts",
@@ -1680,7 +1680,7 @@
     },
     {
       "path": "modules/traceability/package.json",
-      "sha256": "23191c5d6fd9f685376cbb18fbcc8f28fb689dbd258046b4b74f5d8f76817b0c"
+      "sha256": "e06fdfae1aaaaeecf501f077131cba9ad803c720410be8ee919243fbfa5f64aa"
     },
     {
       "path": "modules/traceability/src/contracts.ts",
@@ -1752,7 +1752,7 @@
     },
     {
       "path": "packages/branding/package.json",
-      "sha256": "f351a461d51ad492b725b158da51cc465d54a769f290efd20c896d0c6c7ca159"
+      "sha256": "fcb4dc9e0764e55c03bc4219eb22741120da3d29a013b30016eb3ac68c9ecc55"
     },
     {
       "path": "packages/branding/src/index.ts",
@@ -1764,7 +1764,7 @@
     },
     {
       "path": "packages/contracts/package.json",
-      "sha256": "ec3c28bd8df852391f9f05544bcf97e57185b365f2982b2a5f9965ad85e4724b"
+      "sha256": "576bc337379e42e6eb1b8bf0eaff110581f7c30b300dd49a80f36dac66cc3a69"
     },
     {
       "path": "packages/contracts/src/export.ts",
@@ -1784,7 +1784,7 @@
     },
     {
       "path": "packages/data-export/package.json",
-      "sha256": "ae468b1c4936071fbf57f2b7c6a9c073c39dfa827a9b16e849b74f7a1dd08f56"
+      "sha256": "8a2851680095d19241b5c82ad768af241346ba08e62bc138adaba28e4c3dd730"
     },
     {
       "path": "packages/data-export/src/index.test.ts",
@@ -1808,15 +1808,15 @@
     },
     {
       "path": "packages/local-runtime/dist/server.cjs",
-      "sha256": "6247f923afacbf3b5be41beea52e218385d303ace283d69744c0eeb9afc9917a"
+      "sha256": "8b5a32cf093fa09b7e4fd04b5a64a8162e79221f08fcef1fb5e9ec39dc50a0b1"
     },
     {
       "path": "packages/local-runtime/dist/sqlite-read-model.cjs",
-      "sha256": "9f2f4c64d8b3b803eca23ddf72d90d8b3befe22c710effec8f6d4d4d441a7bdd"
+      "sha256": "8a5a66816c262ce5ef09abfde4c262f2aa63cb869be49df76e7414efcddccb36"
     },
     {
       "path": "packages/local-runtime/package.json",
-      "sha256": "413b1c75aa4c59a0414cdf9bd4e27fa52d9ebaab8ff7b0d5574f3fc9fb45d7fb"
+      "sha256": "0e0c3ec04774d2347a2347cfa05b01de10e3f936a97652f48da7e078f89062bf"
     },
     {
       "path": "packages/local-runtime/scripts/build-sqlite-dashboard.mjs",
@@ -1852,7 +1852,7 @@
     },
     {
       "path": "packages/local-runtime/server.mjs",
-      "sha256": "5daeb27d545afda48158801f1560676526770abc807898868aac27e31010c8c1"
+      "sha256": "11bd0e7f9de4e9bc24353dec3f82a1747560353c5e5625a8a4c8a9adf9d65700"
     },
     {
       "path": "packages/local-runtime/src/active-codex-telemetry.test.ts",
@@ -1974,6 +1974,10 @@
       "path": "packages/local-runtime/src/server-macro-population.test.ts",
       "sha256": "cdfe55242dbeb1d6d0eec87781221ad80325a30fcb84de6924707037c8ba76e5"
     },
+    {
+      "path": "packages/local-runtime/src/server-operation-detail.test.ts",
+      "sha256": "77922d17e7b82cbed608c61b45f34af4f7fd0d8cb2d0ff6c50d441d71ec122ff"
+    },
     {
       "path": "packages/local-runtime/src/server-operations-page-filters.ts",
       "sha256": "26259f9bc22e10a75086f0c520bf5ca2320d4a2e72dfc0770e1d67034b8fb047"
@@ -2128,7 +2132,7 @@
     },
     {
       "path": "packages/local-runtime/src/server.test.ts",
-      "sha256": "6b8d006446755d26ab119745dd4d47407e83bf0cf9c1487c0a2cd16e27534d40"
+      "sha256": "81bfe4dc9bd019a50892e64584ee10bb6ab8e261497e542fde5bd612cbb46b72"
     },
     {
       "path": "packages/local-runtime/src/server.ts",
@@ -2380,7 +2384,11 @@
     },
     {
       "path": "packages/local-runtime/src/sqlite-read-model-counts-every-root-execution-day-returns.test.ts",
-      "sha256": "c3c4bc0ceddcbf0addeae9d64b1b0e9128f6f0ca9ec5aadb618c3c1605b9fd50"
+      "sha256": "754c06e8774234062eaf8b45f7ccc4a6fb745d448aaf82469acef557c7b5c2c6"
+    },
+    {
+      "path": "packages/local-runtime/src/sqlite-read-model-counts-fixture.test-support.ts",
+      "sha256": "782e4571cdfbb332654078b1896e814d85b9deadb351198a7c5f5397d8dae432"
     },
     {
       "path": "packages/local-runtime/src/sqlite-read-model-does-not-convert-delayed-historical-finalizer.test.ts",
@@ -2540,7 +2548,7 @@
     },
     {
       "path": "packages/local-runtime/start-dashboard-launch-page.test.mjs",
-      "sha256": "908ff7972dc0e6d2fc631bd96c7572383e975b8c81ecd5ab711682d85a74e598"
+      "sha256": "ce94603b829d2db9af509fc9d7b63cbe970046580b4a093433043d8db3c97749"
     },
     {
       "path": "packages/local-runtime/start-dashboard-lifecycle.mjs",
@@ -2612,7 +2620,7 @@
     },
     {
       "path": "packages/ui/package.json",
-      "sha256": "4573a6671dc05fa8604e474d2cfb70dc48b0abfb25b94d95e10af127fe3097be"
+      "sha256": "9a9e598e9cc4b9e17ea2b152125c45b4fda7fd0a187261087b09378ea0b45b7a"
     },
     {
       "path": "packages/ui/src/DashboardHeader.test.mjs",
@@ -2720,7 +2728,7 @@
     },
     {
       "path": "tests/e2e/package.json",
-      "sha256": "89247d6597c2e8f32c62dab3065a5bb138b376715baefeeac05e20a40bad996e"
+      "sha256": "8ec1a02b9436f4771e81c8dcc0e8c833549628c2a25cc06be70778f812d86551"
     },
     {
       "path": "tests/e2e/playwright.config.ts",