← Plugin catalog
Business & Operations
Marketing Council
Mamdouh Aboammar v1.5.0
Publisher description
From the marketplace listing
Marketing Council routes marketing briefs across 28 focused skills, with the Council as the 29th skill and the fallback for ambiguous or cross-functional work. It combines specialist agents, evidence checks, commercial constraints, current-state research, and causal measurement discipline.
Language: English · Automatically detected from descriptions.
Files & skills
File archives
Plugin package623 files · 198 KBBrowse files →
Skill instructions
agentic-commerce2.99 KB
--- name: agentic-commerce description: Use when the user needs to prepare products, policies, offers, and authority boundaries for agent-mediated buying; route here only when this decision boundary is the clear owner. --- # Agentic Commerce ## Job Prepare products, policies, offers, and authority boundaries for agent-mediated buying Own this request only when **agentic commerce readiness** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/agentic-commerce-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/agentic-commerce-readiness.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `agentic-commerce` - Decision boundary: `agentic commerce readiness` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
ai-discovery-strategy2.96 KB
--- name: ai-discovery-strategy description: Use when the user needs to improve retrievability and truthful representation across AI answer surfaces; route here only when this decision boundary is the clear owner. --- # Ai Discovery Strategy ## Job Improve retrievability and truthful representation across AI answer surfaces Own this request only when **AI discovery intervention** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/ai-discovery-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/ai-surface-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `ai-discovery-strategy` - Decision boundary: `AI discovery intervention` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
autonomous-media-operations3.02 KB
--- name: autonomous-media-operations description: Use when the user needs to define decision rights, guardrails, monitoring, and rollback for automated media operations; route here only when this decision boundary is the clear owner. --- # Autonomous Media Operations ## Job Define decision rights, guardrails, monitoring, and rollback for automated media operations Own this request only when **automation authority** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/marketing-automation-governor.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/automation-authority-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `autonomous-media-operations` - Decision boundary: `automation authority` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
behavioral-marketing2.97 KB
--- name: behavioral-marketing description: Use when the user needs to use behavioral evidence to reduce friction and shape ethical choice architecture; route here only when this decision boundary is the clear owner. --- # Behavioral Marketing ## Job Use behavioral evidence to reduce friction and shape ethical choice architecture Own this request only when **behavioral intervention** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/behavior-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/causal-mechanism-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `behavioral-marketing` - Decision boundary: `behavioral intervention` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
brand-strategy2.96 KB
--- name: brand-strategy description: Use when the user needs to define durable brand associations, distinctive assets, architecture, and memory priorities; route here only when this decision boundary is the clear owner. --- # Brand Strategy ## Job Define durable brand associations, distinctive assets, architecture, and memory priorities Own this request only when **brand strategy** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/brand-equity-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/horizon-balance-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `brand-strategy` - Decision boundary: `brand strategy` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
campaign-strategy2.95 KB
--- name: campaign-strategy description: Use when the user needs to turn a diagnosed objective into an integrated campaign strategy with one clear job; route here only when this decision boundary is the clear owner. --- # Campaign Strategy ## Job Turn a diagnosed objective into an integrated campaign strategy with one clear job Own this request only when **campaign decision** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/creative-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/anti-generic-marketing.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `campaign-strategy` - Decision boundary: `campaign decision` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
category-strategy2.95 KB
--- name: category-strategy description: Use when the user needs to decide category frame, maturity response, entry points, and demand creation posture; route here only when this decision boundary is the clear owner. --- # Category Strategy ## Job Decide category frame, maturity response, entry points, and demand creation posture Own this request only when **category frame** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/market-architect.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/category-maturity-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `category-strategy` - Decision boundary: `category frame` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
commerce-feed-intelligence2.99 KB
--- name: commerce-feed-intelligence description: Use when the user needs to make commerce feeds accurate, complete, semantically useful, and decision-ready; route here only when this decision boundary is the clear owner. --- # Commerce Feed Intelligence ## Job Make commerce feeds accurate, complete, semantically useful, and decision-ready Own this request only when **feed quality decision** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/agentic-commerce-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/commerce-feed-readiness.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `commerce-feed-intelligence` - Decision boundary: `feed quality decision` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
commerce-media-strategy3.01 KB
--- name: commerce-media-strategy description: Use when the user needs to plan retail and commerce media with shopper roles, economics, and closed-loop measurement safeguards; route here only when this decision boundary is the clear owner. --- # Commerce Media Strategy ## Job Plan retail and commerce media with shopper roles, economics, and closed-loop measurement safeguards Own this request only when **commerce media plan** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/commerce-media-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/closed-loop-bias-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `commerce-media-strategy` - Decision boundary: `commerce media plan` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
competitive-intelligence3.11 KB
--- name: competitive-intelligence description: Use when the user needs to map real competitive alternatives, substitutes, rival moves, and defensibility without collapsing into positioning or generic research; route here only when this decision boundary is the clear owner. --- # Competitive Intelligence ## Job Map real competitive alternatives, substitutes, rival moves, and defensibility without collapsing into positioning or generic research Own this request only when **competitive intelligence assessment** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/competitive-strategy-analyst.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/pre-mortem.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `competitive-intelligence` - Decision boundary: `competitive intelligence assessment` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
content-strategy2.95 KB
--- name: content-strategy description: Use when the user needs to define content jobs, formats, cadence, and distribution tied to audience decisions; route here only when this decision boundary is the clear owner. --- # Content Strategy ## Job Define content jobs, formats, cadence, and distribution tied to audience decisions Own this request only when **content portfolio** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/creative-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/anti-generic-marketing.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `content-strategy` - Decision boundary: `content portfolio` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
conversational-advertising3.01 KB
--- name: conversational-advertising description: Use when the user needs to design ads that use conversation to qualify intent and progress decisions without inventing user context; route here only when this decision boundary is the clear owner. --- # Conversational Advertising ## Job Design ads that use conversation to qualify intent and progress decisions without inventing user context Own this request only when **conversation flow** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/response-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/ai-surface-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `conversational-advertising` - Decision boundary: `conversation flow` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
conversion-strategy2.93 KB
--- name: conversion-strategy description: Use when the user needs to diagnose and reduce friction across high-value conversion paths; route here only when this decision boundary is the clear owner. --- # Conversion Strategy ## Job Diagnose and reduce friction across high-value conversion paths Own this request only when **conversion intervention** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/response-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/causal-mechanism-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `conversion-strategy` - Decision boundary: `conversion intervention` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
creator-commerce3.02 KB
--- name: creator-commerce description: Use when the user needs to design creator roles across discovery, media, affiliate, and commerce with clear economics and measurement; route here only when this decision boundary is the clear owner. --- # Creator Commerce ## Job Design creator roles across discovery, media, affiliate, and commerce with clear economics and measurement Own this request only when **creator commerce program** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/creator-commerce-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/creator-measurement-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `creator-commerce` - Decision boundary: `creator commerce program` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
customer-research2.99 KB
--- name: customer-research description: Use when the user needs to design research that reveals customer progress, switching triggers, language, and decision criteria; route here only when this decision boundary is the clear owner. --- # Customer Research ## Job Design research that reveals customer progress, switching triggers, language, and decision criteria Own this request only when **customer evidence** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/audience-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/customer-language-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `customer-research` - Decision boundary: `customer evidence` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
go-to-market2.94 KB
--- name: go-to-market description: Use when the user needs to sequence market entry, audience, channels, enablement, and launch dependencies; route here only when this decision boundary is the clear owner. --- # Go To Market ## Job Sequence market entry, audience, channels, enablement, and launch dependencies Own this request only when **go-to-market sequence** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/channel-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/strategy-before-tactics.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `go-to-market` - Decision boundary: `go-to-market sequence` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
incrementality-design2.95 KB
--- name: incrementality-design description: Use when the user needs to design counterfactual measurement that estimates causal marketing lift; route here only when this decision boundary is the clear owner. --- # Incrementality Design ## Job Design counterfactual measurement that estimates causal marketing lift Own this request only when **incrementality design** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/measurement-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/incrementality-required.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `incrementality-design` - Decision boundary: `incrementality design` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
market-diagnosis2.92 KB
--- name: market-diagnosis description: Use when the user needs to diagnose the commercial growth constraint before prescribing tactics; route here only when this decision boundary is the clear owner. --- # Market Diagnosis ## Job Diagnose the commercial growth constraint before prescribing tactics Own this request only when **market constraint** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/market-architect.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/strategy-before-tactics.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `market-diagnosis` - Decision boundary: `market constraint` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
marketing-council4.26 KB
--- name: marketing-council description: Use when the user needs to diagnose ambiguous or cross-functional marketing problems and coordinate the minimum set of specialist skills; route here only when this decision boundary is the clear owner. --- # Marketing Council ## Job Diagnose ambiguous or cross-functional marketing problems and coordinate the minimum set of specialist skills Own the request when the marketing problem is ambiguous or genuinely cross-functional. If a single dominant function clearly owns the next decision, delegate to that focused Skill instead of retaining Council ownership. Otherwise Council is the safe fallback. If the request explicitly establishes dependent work across functions, use the dynamic router in `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Canonical Skill Router registry: `../../routing/skill-routes.json`. - Single dominant function: delegate to the selected focused Skill. - Ambiguous or cross-functional ownership: Council remains the fallback diagnostic owner. - Explicit dependency chain: use the dynamic router at `../../scripts/dynamic_router.py` and keep the graph bounded to the minimum required Skills. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/council-director.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/post-strategy-red-team.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Council execution resources - Synthesis and conflict resolution: `../../agents/council-director.md` - Adversarial falsification: `../../agents/marketing-skeptic.md` - Measurement and causality: `../../agents/measurement-strategist.md` - Host capability contract: `../../tools/capabilities.yml` - Focused Skill modules are under `skills/` in the standalone Council bundle; delegate only after ownership is established. ### Principle canon - Market structure and segmentation: `../../references/canon/kotler.md` - Relevance and smallest viable audience: `../../references/canon/godin.md` - Product focus and demonstration: `../../references/canon/jobs-product-principles.md` - Proposition and proof: `../../references/canon/ogilvy.md` - Awareness and sophistication: `../../references/canon/schwartz.md` - Reach, availability, and distinctive assets: `../../references/canon/sharp.md` - Short and long effectiveness horizons: `../../references/canon/binet-field.md` ## Neural connections - Owning Skill: `marketing-council` - Decision boundary: `cross-functional decision` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
marketing-experimentation2.96 KB
--- name: marketing-experimentation description: Use when the user needs to design decision-relevant marketing experiments with falsifiable hypotheses; route here only when this decision boundary is the clear owner. --- # Marketing Experimentation ## Job Design decision-relevant marketing experiments with falsifiable hypotheses Own this request only when **experiment design** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/measurement-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/causal-mechanism-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `marketing-experimentation` - Decision boundary: `experiment design` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
marketing-measurement3.02 KB
--- name: marketing-measurement description: Use when the user needs to build a measurement architecture linking business outcomes, leading indicators, and attribution limits; route here only when this decision boundary is the clear owner. --- # Marketing Measurement ## Job Build a measurement architecture linking business outcomes, leading indicators, and attribution limits Own this request only when **measurement architecture** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/measurement-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/closed-loop-bias-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `marketing-measurement` - Decision boundary: `measurement architecture` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
marketing-signal-strategy2.99 KB
--- name: marketing-signal-strategy description: Use when the user needs to design high-quality optimization signals that reflect customer and business value; route here only when this decision boundary is the clear owner. --- # Marketing Signal Strategy ## Job Design high-quality optimization signals that reflect customer and business value Own this request only when **signal architecture** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/marketing-signal-architect.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/marketing-signal-quality.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `marketing-signal-strategy` - Decision boundary: `signal architecture` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
media-strategy2.95 KB
--- name: media-strategy description: Use when the user needs to assign channel jobs, reach/frequency logic, budget principles, and measurement boundaries; route here only when this decision boundary is the clear owner. --- # Media Strategy ## Job Assign channel jobs, reach/frequency logic, budget principles, and measurement boundaries Own this request only when **media allocation** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/channel-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/horizon-balance-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `media-strategy` - Decision boundary: `media allocation` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
offer-strategy2.93 KB
--- name: offer-strategy description: Use when the user needs to design an offer whose value, proof, risk, and terms fit the buying decision; route here only when this decision boundary is the clear owner. --- # Offer Strategy ## Job Design an offer whose value, proof, risk, and terms fit the buying decision Own this request only when **offer architecture** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/response-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/commercial-reality-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `offer-strategy` - Decision boundary: `offer architecture` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
positioning-strategy2.98 KB
--- name: positioning-strategy description: Use when the user needs to choose a differentiated competitive frame and reason to choose based on real alternatives; route here only when this decision boundary is the clear owner. --- # Positioning Strategy ## Job Choose a differentiated competitive frame and reason to choose based on real alternatives Own this request only when **positioning choice** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/positioning-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/category-maturity-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `positioning-strategy` - Decision boundary: `positioning choice` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
pricing-strategy2.97 KB
--- name: pricing-strategy description: Use when the user needs to set pricing architecture, tiers, and discount guardrails from willingness to pay and economics; route here only when this decision boundary is the clear owner. --- # Pricing Strategy ## Job Set pricing architecture, tiers, and discount guardrails from willingness to pay and economics Own this request only when **pricing decision** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/commercial-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/commercial-reality-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `pricing-strategy` - Decision boundary: `pricing decision` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
product-marketing2.99 KB
--- name: product-marketing description: Use when the user needs to translate product truth into audience-specific narrative, proof, launch messaging, and enablement; route here only when this decision boundary is the clear owner. --- # Product Marketing ## Job Translate product truth into audience-specific narrative, proof, launch messaging, and enablement Own this request only when **product narrative** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/product-marketing-director.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/customer-language-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `product-marketing` - Decision boundary: `product narrative` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
retention-strategy2.95 KB
--- name: retention-strategy description: Use when the user needs to diagnose churn and repeat behavior, then design lifecycle interventions; route here only when this decision boundary is the clear owner. --- # Retention Strategy ## Job Diagnose churn and repeat behavior, then design lifecycle interventions Own this request only when **retention intervention** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/lifecycle-strategist.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/commercial-reality-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `retention-strategy` - Decision boundary: `retention intervention` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
segmentation-strategy2.95 KB
--- name: segmentation-strategy description: Use when the user needs to create commercially useful segments and targeting priorities that change decisions; route here only when this decision boundary is the clear owner. --- # Segmentation Strategy ## Job Create commercially useful segments and targeting priorities that change decisions Own this request only when **segment choice** is the clear decision boundary. If ownership is ambiguous or several functions compete, route to `marketing-council`. If the user explicitly asks for dependent work across functions, use `../../scripts/dynamic_router.py` to build a bounded DAG. ## Operating contract 1. Read `references/skill-spec.json` first for activation, invariants, workflow freedom, evidence rules, handoffs, and completion conditions. 2. Use `references/decision-model.md` when framing or challenging the decision. 3. Check `references/failure-modes.md` before finalizing a recommendation. 4. Render the response against `references/output-contract.md`. 5. Use packaged shared references or current external research only when they are load-bearing. Never present inference as evidence. ## Evidence discipline Classify material claims as fact, inference, assumption, or unknown. Prefer supplied primary evidence. Verify current platform, policy, product, pricing, or market claims when freshness affects the recommendation. Do not fabricate research, tool calls, metrics, customer language, or causal proof. ## Routing - Focused request: stay inside this Skill. - Ambiguous or cross-functional ownership: hand to `marketing-council`. - Explicit dependency chain: use `../../scripts/dynamic_router.py`. - After Skill ownership is known, theory/agent selection may use `../../scripts/neural_router.py`; neural nodes never replace Skill routing. ## Execution connections - Primary specialist: `../../agents/market-architect.md` - Skeptical counterweight: `../../agents/marketing-skeptic.md` - Domain challenge gate: `../../hooks/decision-stage-check.md` - Evidence gate: `../../hooks/evidence-gate.md` - Keep these as decision inputs, not automatic authority. The Skill owns the final evidence-bound synthesis. ## Neural connections - Owning Skill: `segmentation-strategy` - Decision boundary: `segment choice` - Neural graph: `../../neural/graph.json` - Neural router: `../../scripts/neural_router.py` - Theory and specialist selection happens only after Skill ownership; neural nodes never replace Skill routing. - Use the local `references/skill-spec.json` evidence policy and invariants to reject neural recommendations that are unsupported by the request evidence. ## Completion gate Complete only when the decision is explicit, evidence and inference are separated, a credible alternative was considered, outputs are rendered, material uncertainty is stated, and measurement plus reversal evidence are defined. Local behavioral evaluations live in `evals/activation.yml`, `evals/behavior.yml`, `evals/pressure.yml`, and `evals/regression.yml`.
Referenced files: 9
Package details
Publisher declarations from the archived package. These are separate from our research and the live service's terms.
- Package license
- MIT
- Package author
- Mamdouh Aboammar
- Keywords
- marketing, strategy, positioning, go-to-market, pricing, campaigns, cro, retention
Declared capabilities
- Market diagnosis
- Customer research
- Segmentation and targeting
- Positioning and category strategy
- Brand strategy
- Product marketing
- Offer and pricing strategy
- Go-to-market planning
- Campaign and creative strategy
- Media and channel strategy
- Behavioral marketing and CRO
- Retention and lifecycle
- Competitive intelligence
- Marketing measurement
- Marketing experimentation
- Cross-functional strategy arbitration
- AI-mediated discovery strategy
- Agentic commerce and commerce feeds
- Autonomous media governance
- Incrementality and causal measurement
Some manifest fields differ or could not be read. The structured report retains the source references.
Package observed Oct 2, 2026.
Technical details
- First seen
- Sep 30, 2026 · 22:02 UTC
- Last seen
- Oct 2, 2026 · 12:00 UTC
- Collection status
- Collected
plugins_6a8366789d3081919bf20654f87e082b
Download plugin data (JSON)