@monoes/monomindcli 1.8.0 → 1.9.1

package/.claude/skills/mastermind/finance.md

@@ -0,0 +1,154 @@
+---
+name: mastermind-finance
+description: Mastermind finance domain — invoicing, budget tracking, financial forecasting. Spawns a Finance Manager coordinating financial tracking and analysis agents.
+type: domain-skill
+default_mode: confirm
+---
+
+# Mastermind Finance Domain
+
+This skill is invoked by `mastermind:master` or directly via `/mastermind:finance`.
+
+---
+
+## Inputs
+
+- `brain_context`: BRAIN CONTEXT block (injected by master, or loaded standalone via _protocol.md brain load)
+- `prompt`: the finance goal for this run
+- `project_name`: monotask space name
+- `board_id`: monotask board ID (set by master, or created standalone)
+- `mode`: auto | confirm
+
+---
+
+## Complexity Assessment
+
+Assess the prompt to determine execution mode:
+
+**Simple (direct execution):** Single calculation or single document:
+- "Calculate total revenue for Q1 from these figures: ..."
+- "Draft an invoice for project X at $5,000"
+→ Use a single Finance Tracker agent. Skip manager delegation.
+
+**Complex (spawn Finance Manager agent):** Any of these:
+- Full financial forecast (revenue + costs + runway)
+- Budget analysis across multiple cost centers
+- Batch invoice processing
+- P&L or cash flow modeling
+→ Spawn Finance Manager agent with full briefing.
+
+---
+
+## Standalone Execution (when called without master)
+
+If this skill is invoked directly (not by master):
+
+1. Load brain context following _protocol.md Brain Load Procedure (namespace: `finance`)
+2. Run intake from _intake.md if prompt is vague
+3. Create or find monotask space `<project_name>`, create board `finance`
+4. Proceed with complexity assessment below
+5. At end: follow _protocol.md Brain Write Procedure (namespace: `finance`)
+
+---
+
+## Complex Execution — Finance Manager Agent
+
+Spawn a Finance Manager agent via Task tool:
+
+```javascript
+Task({
+  subagent_type: "coordinator",
+  description: `You are the Finance Manager for project <project_name>.
+
+CONTEXT: <date> | Project: <project_name> | Spawned by: mastermind:finance
+
+BRAIN CONTEXT:
+<brain_context>
+
+YOUR BOARD: <board_id>
+YOUR GOAL: <prompt>
+
+STEP 1 — PLAN
+Decompose the finance goal into discrete workstreams. For each workstream, identify:
+- What financial domain it covers (invoicing, tracking, forecasting, analysis)
+- What input data is needed
+- What deliverable is produced (report, model, invoice batch)
+- Dependencies between workstreams
+
+STEP 2 — CREATE TASKS
+For each workstream, create a monotask card on the project board. First look up column IDs and assign shell variables:
+```bash
+columns=$(monotask column list "$BOARD_ID" --json)
+COL_TODO_ID=$(echo "$columns" | jq -r '.[] | select(.name == "Todo" or .name == "Backlog") | .id' | head -1)
+COL_DONE_ID=$(echo "$columns" | jq -r '.[] | select(.name == "Done") | .id' | head -1)
+```
+Then create the card:
+```bash
+result=$(monotask card create "$BOARD_ID" "$COL_TODO_ID" "<short summary of workstream goal, ≤80 chars>" --json)
+CARD_ID=$(echo "$result" | jq -r '.id // empty')
+monotask card set-description "$BOARD_ID" "$CARD_ID" "[specific financial workstream goal]"
+monotask card comment add "$BOARD_ID" "$CARD_ID" "CONTEXT: <date> | Project: <project_name> | Created by: Finance Manager
+BRAIN MEMORY: [paste most relevant 3-5 brain context excerpts]
+SCOPE: [time period, accounts, cost centers, currencies in scope]
+CONSTRAINTS: [accounting standards, tax requirements, rounding rules, approval thresholds]
+SUCCESS CRITERIA:
+- [ ] [checkable item — e.g. \"forecast covers 12-month runway\"]
+AGENT: [Finance Tracker | Analytics Reporter]
+SWARM: hierarchical 3 raft
+DEPENDENCIES: [task IDs or \"none\"]
+OUTPUT FORMAT: unified output schema"
+```
+
+STEP 3 — EXECUTE
+Spawn one Task agent per workstream:
+- Financial tracking and invoicing: subagent_type "Finance Tracker"
+- Analysis and reporting: subagent_type "Analytics Reporter"
+
+Also run /monomind:do --board <board_id> to track execution.
+
+STEP 4 — COLLECT AND RETURN
+Collect all financial outputs. Return to caller:
+
+domain: finance
+status: complete | partial | blocked
+artifacts:
+  - path: [financial reports, invoices, forecast models]
+    type: report
+decisions:
+  - what: [financial decisions or recommendations made]
+    why: [reasoning — cost drivers, runway implications, risk factors]
+    confidence: [0.0-1.0]
+    outcome: pending | shipped
+lessons:
+  - what_worked: [what data sources or models were most accurate]
+  - what_didnt: [what required manual correction or more data]
+next_actions:
+  - [e.g. "run mastermind:ops to automate invoice generation"]
+  - [e.g. "run mastermind:review on the financial model assumptions"]
+board_url: monotask://<project_name>/finance
+run_id: <ISO8601-timestamp>`,
+  run_in_background: true
+})
+```
+
+---
+
+## Simple Execution
+
+For simple tasks (single agent, single output):
+
+1. Spawn one Task agent with the finance request as a self-contained briefing
+2. Collect output
+3. Return unified output schema with `status: complete`
+
+---
+
+## Domain Swarm Defaults
+
+| Task Type | Agent | Swarm |
+|---|---|---|
+| Full financial forecast | coordinator + tracker + analyst | hierarchical 3 raft specialized |
+| Budget analysis | Finance Tracker + Analytics Reporter | hierarchical 3 raft specialized |
+| Invoice batch | Finance Tracker | hierarchical 3 raft specialized |
+| P&L report | Analytics Reporter | single agent |
+| Single invoice or calculation | Finance Tracker | single agent |

package/.claude/skills/mastermind/idea.md

@@ -0,0 +1,158 @@
+---
+name: mastermind-idea
+description: Mastermind idea domain — product ideation, feature brainstorming, pivot exploration. Spawns an Idea Manager agent who coordinates a mesh of specialist agents for divergent thinking.
+type: domain-skill
+default_mode: confirm
+---
+
+# Mastermind Idea Domain
+
+This skill is invoked by `mastermind:master` or directly via `/mastermind:idea`.
+
+---
+
+## Inputs
+
+- `brain_context`: BRAIN CONTEXT block (injected by master, or loaded standalone via _protocol.md brain load)
+- `prompt`: the ideation goal for this run
+- `project_name`: monotask space name
+- `board_id`: monotask board ID (set by master, or created standalone)
+- `mode`: auto | confirm
+
+---
+
+## Complexity Assessment
+
+Assess the prompt to determine execution mode:
+
+**Simple (direct execution):** Single-answer ideation, one agent:
+- "Give me 5 name ideas for this feature"
+- "Suggest one pivot angle for this product"
+→ Use a single researcher or content-creator agent. Skip manager delegation.
+
+**Complex (spawn Idea Manager agent):** Any of these:
+- Product strategy or pivot exploration (multiple angles needed)
+- Feature ideation requiring market and user context
+- Competitive landscape brainstorm
+- Full product vision document
+→ Spawn Idea Manager agent with full briefing.
+
+---
+
+## Standalone Execution (when called without master)
+
+If this skill is invoked directly (not by master):
+
+1. Load brain context following _protocol.md Brain Load Procedure (namespace: `idea`)
+2. Run intake from _intake.md if prompt is vague
+3. Create or find monotask space `<project_name>`, create board `ideation`
+4. Proceed with complexity assessment below
+5. At end: follow _protocol.md Brain Write Procedure (namespace: `idea`)
+
+---
+
+## Complex Execution — Idea Manager Agent
+
+Spawn an Idea Manager agent via Task tool:
+
+```javascript
+Task({
+  subagent_type: "coordinator",
+  description: `You are the Idea Manager for project <project_name>.
+
+CONTEXT: <date> | Project: <project_name> | Spawned by: mastermind:idea
+
+BRAIN CONTEXT:
+<brain_context>
+
+YOUR BOARD: <board_id>
+YOUR GOAL: <prompt>
+
+STEP 1 — PLAN
+Decompose the ideation goal into distinct angles of exploration. For each angle, identify:
+- What perspective or lens to apply (market, user, technical, competitive)
+- Which specialist to assign
+- What output format is needed (list, document, diagram, recommendation)
+- Dependencies between angles
+
+STEP 2 — CREATE TASKS
+For each angle, create a monotask card on the project board.
+First look up column IDs and assign shell variables:
+```bash
+columns=$(monotask column list "$BOARD_ID" --json)
+COL_TODO_ID=$(echo "$columns" | jq -r '.[] | select(.name == "Todo" or .name == "Backlog") | .id' | head -1)
+COL_DONE_ID=$(echo "$columns" | jq -r '.[] | select(.name == "Done") | .id' | head -1)
+```
+Then create the card:
+```bash
+result=$(monotask card create "$BOARD_ID" "$COL_TODO_ID" "<short summary of ideation angle, ≤80 chars>" --json)
+CARD_ID=$(echo "$result" | jq -r '.id // empty')
+monotask card set-description "$BOARD_ID" "$CARD_ID" "[specific ideation angle and output]"
+monotask card comment add "$BOARD_ID" "$CARD_ID" "CONTEXT: <date> | Project: <project_name> | Created by: Idea Manager
+BRAIN MEMORY: [paste most relevant 3-5 brain context excerpts]
+SCOPE: [domain of exploration — market, users, technology, competitors]
+CONSTRAINTS: [existing product constraints, brand voice, strategic limits]
+SUCCESS CRITERIA:
+- [ ] [checkable item]
+AGENT: [researcher | Trend Researcher | Product Manager | Growth Hacker | Content Creator]
+SWARM: mesh 6 gossip
+DEPENDENCIES: [task IDs or \"none\"]
+OUTPUT FORMAT: unified output schema"
+```
+
+STEP 3 — EXECUTE
+Spawn one Task agent per angle (all in parallel — mesh topology means all perspectives feed each other):
+- Market research angle: subagent_type "researcher"
+- Trend analysis: subagent_type "Trend Researcher"
+- User perspective: subagent_type "UX Researcher"
+- Growth angle: subagent_type "Growth Hacker"
+- Content/narrative angle: subagent_type "Content Creator"
+
+Also run /monomind:do --board <board_id> to track execution.
+
+STEP 4 — COLLECT AND RETURN
+Synthesize all agent perspectives into a coherent ideation report. Return to caller:
+
+domain: idea
+status: complete | partial | blocked
+artifacts:
+  - path: [ideation document if written to disk]
+    type: report
+decisions:
+  - what: [top idea or direction recommended]
+    why: [reasoning from synthesized perspectives]
+    confidence: [0.0-1.0]
+    outcome: pending
+lessons:
+  - what_worked: [which angles produced the best insights]
+  - what_didnt: [which angles were less useful]
+next_actions:
+  - [e.g. "run mastermind:research to validate top idea"]
+  - [e.g. "run mastermind:build to prototype chosen direction"]
+board_url: monotask://<project_name>/ideation
+run_id: <ISO8601-timestamp>`,
+  run_in_background: true
+})
+```
+
+---
+
+## Simple Execution
+
+For simple tasks (single agent, single output):
+
+1. Spawn one Task agent with the ideation request as a self-contained briefing
+2. Collect output
+3. Return unified output schema with `status: complete`
+
+---
+
+## Domain Swarm Defaults
+
+| Task Type | Agent | Swarm |
+|---|---|---|
+| Full product ideation | coordinator + mesh specialists | mesh 6 gossip balanced |
+| Feature brainstorm | researcher + trend-analyst | mesh 4 gossip balanced |
+| Pivot exploration | coordinator + researcher + growth | mesh 4 gossip balanced |
+| Competitive scan | researcher | hierarchical 3 raft specialized |
+| Single idea request | researcher or content-creator | single agent |

package/.claude/skills/mastermind/marketing.md

@@ -0,0 +1,216 @@
+---
+name: mastermind-marketing
+description: Mastermind marketing domain — campaigns, copy, SEO, social media strategy. Spawns a Marketing Manager who coordinates content, SEO, social, and analytics agents in parallel.
+type: domain-skill
+default_mode: confirm
+---
+
+# Mastermind Marketing Domain
+
+This skill is invoked by `mastermind:master` or directly via `/mastermind:marketing`.
+
+---
+
+## Inputs
+
+- `brain_context`: BRAIN CONTEXT block (injected by master, or loaded standalone via _protocol.md brain load)
+- `prompt`: the marketing goal for this run
+- `project_name`: monotask space name
+- `board_id`: monotask board ID (set by master, or created standalone)
+- `mode`: auto | confirm
+
+---
+
+## Reference Library
+
+Before building any task briefing, identify which reference files apply and include their paths in the agent's instructions with a directive to Read them before starting work.
+
+| Reference | When to include |
+|---|---|
+| `references/persuasion-psychology.md` | **Always** — all marketing tasks |
+| `references/copywriting-frameworks.md` | Copy tasks: landing pages, headlines, CTAs, hero sections, any page copy; email copy: subject lines, cold outreach, nurture sequences |
+| `.claude/skills/stop-slop/SKILL.md` | **All copy review** — run before delivering any written asset |
+| `.claude/skills/stop-slop/references/phrases.md` | Full copy edit pass — check for banned phrases |
+| `.claude/skills/stop-slop/references/structures.md` | Full copy edit pass — check structural patterns |
+
+In each agent task briefing, add:
+```
+REFERENCE FILES: Read these before starting:
+- .claude/skills/mastermind/references/persuasion-psychology.md
+[For copy tasks (landing pages, CTAs, email copy, headlines, subject lines), also read:]
+- .claude/skills/mastermind/references/copywriting-frameworks.md
+[For all copy review/edit passes, also read:]
+- .claude/skills/stop-slop/SKILL.md
+- .claude/skills/stop-slop/references/phrases.md
+- .claude/skills/stop-slop/references/structures.md
+```
+
+---
+
+## Specialist Agent Routing
+
+Use these agents instead of generic types when the task is domain-specific:
+
+| Task | Agent slug | When to use |
+|---|---|---|
+| Email sequences, drip campaigns, cold outreach | `email-specialist` | Any multi-email flow or cold email |
+| Page CRO, signup flow, form, popup optimization | `cro-specialist` | Conversion analysis or optimization |
+| Pricing decisions, tier design, pricing pages | `pricing-strategist` | Any pricing or packaging task |
+| Product launch, feature announcement, GTM | `launch-strategist` | Any launch or release |
+| Competitor pages, content strategy, topic planning | `competitive-content` | Competitive content or editorial planning |
+| General copy and content | `content-creator` | Multi-platform content, brand storytelling |
+| SEO audits, keyword strategy | `seo-specialist` | Technical SEO, organic search |
+| Social media | `social-media-strategist` | Social content and calendars |
+| Paid ads | `ad-creative-strategist` | Ad copy and paid creative |
+
+For tasks outside these domains (referral programs, schema markup, analytics tracking), name the appropriate specialist agent in the briefing and let it handle depth.
+
+---
+
+## Complexity Assessment
+
+Assess the prompt to determine execution mode:
+
+**Simple (direct execution):** Single asset, single agent:
+- "Write one tweet announcing the new API release"
+- "Draft a subject line for the newsletter"
+→ Use a single Content Creator agent. Skip manager delegation.
+
+**Complex (spawn Marketing Manager agent):** Any of these:
+- Full campaign across multiple channels
+- SEO audit or keyword strategy
+- Social media calendar or multi-post series
+- Launch strategy requiring copy + analytics + distribution
+→ Spawn Marketing Manager agent with full briefing.
+
+---
+
+## Standalone Execution (when called without master)
+
+If this skill is invoked directly (not by master):
+
+1. Load brain context following _protocol.md Brain Load Procedure (namespace: `marketing`)
+2. Run intake from _intake.md if prompt is vague
+3. Create or find monotask space `<project_name>`, create board `marketing`
+4. Proceed with complexity assessment below
+5. At end: follow _protocol.md Brain Write Procedure (namespace: `marketing`)
+
+---
+
+## Complex Execution — Marketing Manager Agent
+
+Spawn a Marketing Manager agent via Task tool:
+
+```javascript
+Task({
+  subagent_type: "coordinator",
+  description: `You are the Marketing Manager for project <project_name>.
+
+CONTEXT: <date> | Project: <project_name> | Spawned by: mastermind:marketing
+
+BRAIN CONTEXT:
+<brain_context>
+
+YOUR BOARD: <board_id>
+YOUR GOAL: <prompt>
+
+STEP 1 — PLAN
+Decompose the marketing goal into parallel workstreams. For each workstream, identify:
+- Which channel or medium it targets (social, SEO, email, paid, content)
+- Which specialist to assign
+- What deliverables are needed (copy, strategy doc, keyword list, calendar)
+- Dependencies between workstreams
+
+STEP 2 — CREATE TASKS
+For each workstream, create a monotask card on the project board.
+First look up column IDs and assign shell variables:
+```bash
+columns=$(monotask column list "$BOARD_ID" --json)
+COL_TODO_ID=$(echo "$columns" | jq -r '.[] | select(.title == "Todo" or .title == "Backlog") | .id' | head -1)
+COL_DONE_ID=$(echo "$columns" | jq -r '.[] | select(.title == "Done") | .id' | head -1)
+```
+Then create the card:
+```bash
+result=$(monotask card create "$BOARD_ID" "$COL_TODO_ID" "<short summary of workstream goal, ≤80 chars>" --json)
+CARD_ID=$(echo "$result" | jq -r '.id // empty')
+monotask card set-description "$BOARD_ID" "$CARD_ID" "[specific marketing workstream goal]"
+monotask card comment add "$BOARD_ID" "$CARD_ID" "CONTEXT: <date> | Project: <project_name> | Created by: Marketing Manager
+BRAIN MEMORY: [paste most relevant 3-5 brain context excerpts]
+SCOPE: [channel, audience, tone, brand constraints]
+CONSTRAINTS: [brand voice, legal/compliance, budget limits, existing assets]
+REFERENCE FILES: [for copy/content workstreams: .claude/skills/mastermind/references/persuasion-psychology.md; for copy tasks also: .claude/skills/mastermind/references/copywriting-frameworks.md; for copy review/edit: .claude/skills/stop-slop/SKILL.md, .claude/skills/stop-slop/references/phrases.md, .claude/skills/stop-slop/references/structures.md]
+SUCCESS CRITERIA:
+- [ ] [checkable item]
+AGENT: [Content Creator | Email Marketing Specialist | CRO Specialist | Pricing Strategist | Launch Strategist | Competitive Content Strategist | SEO Specialist | Social Media Strategist | Analytics Reporter | Ad Creative Strategist]
+SWARM: star 5 parallel
+DEPENDENCIES: [task IDs or \"none\"]
+OUTPUT FORMAT: unified output schema"
+```
+
+STEP 3 — EXECUTE
+Spawn one Task agent per workstream (all in parallel — star topology, hub aggregates):
+- Copy and content: subagent_type "Content Creator"
+- SEO: subagent_type "SEO Specialist"
+- Social media: subagent_type "Social Media Strategist"
+- Analytics and measurement: subagent_type "Analytics Reporter"
+- Ad creative: subagent_type "Ad Creative Strategist"
+- Email sequences / drip campaigns: subagent_type "Email Marketing Specialist" (slug: email-specialist)
+- Page / flow CRO: subagent_type "CRO Specialist" (slug: cro-specialist)
+- Pricing strategy / packaging: subagent_type "Pricing Strategist" (slug: pricing-strategist)
+- Product launch / GTM: subagent_type "Launch Strategist" (slug: launch-strategist)
+- Competitive content / comparison pages: subagent_type "Competitive Content Strategist" (slug: competitive-content)
+
+Also run /monomind:do --board <board_id> to track execution.
+
+STEP 4 — COLLECT AND RETURN
+Collect all agent outputs. Return to caller:
+
+domain: marketing
+status: complete | partial | blocked
+artifacts:
+  - path: [each asset created — copy doc, keyword list, social calendar]
+    type: copy
+decisions:
+  - what: [channel prioritization or messaging decisions]
+    why: [reasoning]
+    confidence: [0.0-1.0]
+    outcome: pending | shipped
+lessons:
+  - what_worked: [which channels or approaches were strongest]
+  - what_didnt: [what needed more brand context or iteration]
+next_actions:
+  - [e.g. "run mastermind:content to produce the blog series"]
+  - [e.g. "run mastermind:review on the campaign copy"]
+board_url: monotask://<project_name>/marketing
+run_id: <ISO8601-timestamp>`,
+  run_in_background: true
+})
+```
+
+---
+
+## Simple Execution
+
+For simple tasks (single agent, single asset):
+
+1. Spawn one Task agent with the marketing request as a self-contained briefing
+2. Collect output
+3. Return unified output schema with `status: complete`
+
+---
+
+## Domain Swarm Defaults
+
+| Task Type | Agent | Swarm |
+|---|---|---|
+| Full campaign | coordinator + channel specialists | star 5 parallel |
+| SEO strategy | SEO Specialist | hierarchical 3 raft specialized |
+| Social calendar | Social Media Strategist | hierarchical 3 raft specialized |
+| Copy production | Content Creator | single agent or hierarchical 3 |
+| Email sequence | Email Marketing Specialist | single agent |
+| Cold outreach | Email Marketing Specialist | single agent |
+| Page / flow CRO | CRO Specialist | single agent |
+| Pricing design | Pricing Strategist | single agent |
+| Product launch | Launch Strategist + Content Creator | hierarchical 3 |
+| Competitive content | Competitive Content Strategist | single agent |
+| Analytics review | Analytics Reporter | single agent |