npm - @pennyfarthing/core - Versions diffs - 7.0.1 → 7.4.0 - Mend

@pennyfarthing/core 7.0.1 → 7.4.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (692) hide show

package/pennyfarthing-dist/guides/shared-context.md DELETED Viewed

@@ -1,147 +0,0 @@
-# Shared Context (All Agents)
-**This file provides project context that all agents need.**
-Loaded automatically by `/prime --agent <name>` on agent activation.
----
-## Project Info
-**Project:** Pennyfarthing - Claude Code agent orchestration framework
-**Version:** 7.0.0
-**Type:** ES module with TypeScript (pnpm monorepo)
-**Node:** >=18.0.0
-### Directory Structure
-```
-pennyfarthing-dist/      # Single source of truth for all definitions
-├── agents/              # 10 main agents + 14 official subagents
-├── commands/            # 42 slash commands
-├── guides/              # Behavior guides and patterns
-├── skills/              # 18+ knowledge domain skills
-├── personas/            # 102 themed agent personas
-│   └── themes/          # Theme YAML files
-└── scripts/             # Utility scripts
-packages/                # Monorepo packages
-├── core/                # CLI commands (init, update, doctor)
-├── cyclist/             # Visual terminal (Electron/Express)
-└── shared/              # Shared utilities (skill search, etc.)
-src/                     # Main TypeScript source
-sprint/                  # Sprint tracking (current-sprint.yaml)
-.session/                # Active work sessions
-.claude/                 # Project's Pennyfarthing setup (symlinks)
-```
----
-## Git Strategy
-### Branch Structure
-| Branch | Purpose |
-|--------|---------|
-| `main` | Production-ready code |
-| `develop` | Integration branch |
-| `feat/{story-id}-*` | Feature branches |
-| `fix/{story-id}-*` | Bug fix branches |
-### PR Flow
-1. Create branch from `develop`
-2. Implement changes
-3. PR targets `develop`
-4. After approval, merge to `develop`
-5. Release: merge `develop` to `main`
-### Commit Format
-```
-<type>(<scope>): <subject>
-<body>
-Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
-```
-Types: `feat`, `fix`, `refactor`, `test`, `docs`, `chore`
----
-## Sprint System
-### Files
-| File | Purpose |
-|------|---------|
-| `sprint/current-sprint.yaml` | Active sprint status |
-| `sprint/archive/` | Completed sprints |
-| `sprint/context/` | Epic technical context |
-### Sprint Status Fields
-```yaml
-sprint:
-  number: N
-  goal: "Sprint goal"
-summary:
-  completed_points: N
-  total_points: N
-epics:
-  - id: N
-    title: "Epic title"
-    stories: [...]
-```
----
-## Session System
-### Active Work
-Session files in `.session/{story-id}-session.md` track:
-- Story context and ACs
-- Current phase (sm/tea/dev/review)
-- Workflow status
-- Agent assessments
-### Workflow Phases
-```
-SM (setup) → TEA (tests) → Dev (impl) → Reviewer (approve) → SM (finish)
-```
----
-## Build Commands
-```bash
-npm run build     # TypeScript compilation
-npm run dev       # Watch mode
-npm run clean     # Remove dist/
-npm test          # Node.js native test runner
-npm run lint      # ESLint
-```
----
-## Persona System
-Configured in `.pennyfarthing/config.local.yaml`. Current theme provides character mappings for each agent with style attributes.
-**102 themes available** across categories:
-- **TV Series:** `mash`, `star-trek`, `the-office`, `breaking-bad`, `game-of-thrones`, etc.
-- **Film:** `princess-bride`, `lord-of-the-rings`, `matrix`, `pulp-fiction`, etc.
-- **Literature:** `alice-in-wonderland`, `discworld`, `shakespeare`, `jane-austen`, etc.
-- **Anime:** `one-piece`, `naruto`, `dragon-ball-z`, `death-note`, etc.
-- **Games:** `zelda`, `final-fantasy`, `mass-effect`, `portal`, etc.
-- **History/Mythology:** `greek-mythology`, `norse-mythology`, `ancient-rome`, etc.
-Use `/list-themes` to see all available themes, `/show-theme <name>` for details.
----
-**This context is automatically loaded. Focus on your agent-specific responsibilities.**

package/pennyfarthing-dist/guides/strategic-agent-behavior.md DELETED Viewed

@@ -1,348 +0,0 @@
-# Strategic Agent Behavior (Shared)
-**This file defines common behavior for strategic agents (Orchestrator, PM, SM, Architect, DevOps).**
-**Inherits from:** `shared-agent-behavior.md` - load that first.
-Strategic agents have full project scope. They make cross-repo decisions and coordinate work.
----
-## Strategic Agents
-| Agent | Role | Primary Focus |
-|-------|------|---------------|
-| **Orchestrator** | Meta operations | Process improvement, agent coordination |
-| **PM** | Product Manager | Sprint planning, prioritization, roadmap |
-| **SM** | Scrum Master | Story creation, technical context, finish |
-| **Architect** | System Architect | Design decisions, patterns, consistency |
-| **DevOps** | Infrastructure | CI/CD, deployment, monitoring |
----
-## Context Loading
-Strategic agents load FULL project context on activation:
-```bash
-cd $CLAUDE_PROJECT_DIR
-echo "=== Loading Strategic Context ==="
-# 1. Sprint status (full)
-cat sprint/current-sprint.yaml 2>/dev/null || echo "No sprint configured"
-# 2. Active work sessions
-ls -la .session/*-session.md 2>/dev/null
-# 3. API context (if exists)
-cat API/.claude/context.md 2>/dev/null | head -50
-# 4. UI context (if exists)
-cat UI/.claude/context.md 2>/dev/null | head -50
-# 5. Git status across repos
-echo "=== Git Status ==="
-git status --short --branch
-[ -d "API" ] && echo "--- API ---" && cd API && git status --short --branch && cd ..
-[ -d "UI" ] && echo "--- UI ---" && cd UI && git status --short --branch && cd ..
-```
-### Context Budget
-Strategic agents have a larger context budget (~500-800 lines):
-```
-Agent file:           200-300 lines
-Sprint status:        100-150 lines
-API context:          30-50 lines
-UI context:           30-50 lines
-Active work:          50-100 lines
-Sidecar:              50-100 lines
-----------------------------
-Total:                460-750 lines
-```
----
-## Cross-Repo Awareness
-Strategic agents see and coordinate across ALL repos:
-### Checking Status
-```bash
-# All repos at once
-for repo in . API UI; do
-    [ -d "$repo" ] || continue
-    echo "=== $repo ==="
-    cd "$CLAUDE_PROJECT_DIR/$repo"
-    git branch --show-current
-    git status --short
-    git log --oneline -3
-done
-```
-### Making Cross-Repo Decisions
-When a decision affects multiple repos:
-1. Consider impact on each repo
-2. Document decision in appropriate sidecar
-3. Update shared documentation if needed
-4. Communicate to affected tactical agents
----
-## Planning Workflows
-Strategic agents focus on planning, not implementation:
-### PM Workflows
-- Sprint planning
-- Backlog prioritization
-- Epic creation
-- Roadmap updates
-### SM Workflows
-- Story creation with technical context
-- Story-to-agent handoff
-- Finish-story bookkeeping
-- Sprint tracking
-### Architect Workflows
-- Design decisions
-- Pattern definition
-- Technical spikes
-- Cross-repo consistency
-### DevOps Workflows
-- Infrastructure changes
-- CI/CD updates
-- Deployment planning
-- Monitoring setup
-### Orchestrator Workflows
-- Process improvement
-- Agent coordination
-- Skill maintenance
-- Retrospectives
----
-## PM ↔ Architect Coordination
-Strategic agents collaborate on cross-cutting concerns. PM and Architect have a specific coordination pattern:
-### When PM Requests Architecture Review
-| Trigger | Action |
-|---------|--------|
-| New epic with technical uncertainty | PM → Architect: "Review feasibility" |
-| Story requires new patterns | PM → Architect: "Recommend approach" |
-| Cross-repo changes needed | PM → Architect: "Design integration" |
-| Performance/security concerns | PM → Architect: "Assess risk" |
-### How Architect Responds
-1. **Sync Review (< 1 story point impact):**
-   - Architect reviews in current session
-   - Provides recommendation immediately
-   - PM proceeds with planning
-2. **Async Review (> 1 story point impact):**
-   - Architect creates spike story
-   - Documents findings in `.claude/project/agents/architect-sidecar/`
-   - PM waits for spike completion before finalizing epic
-### Design Approval Flow
-```
-PM identifies need → Architect reviews → Architect recommends
-                                              ↓
-                              ┌───────────────┴───────────────┐
-                              ↓                               ↓
-                         APPROVED                         NEEDS WORK
-                              ↓                               ↓
-                    PM proceeds with epic            Architect documents concerns
-                                                              ↓
-                                                    PM adjusts scope/approach
-                                                              ↓
-                                                    Re-submit for review
-```
-### Handling Disagreements
-If PM and Architect disagree on approach:
-1. Document both perspectives in architect-sidecar
-2. Escalate to Orchestrator for tiebreaker
-3. Orchestrator decision is final for this sprint
-4. Revisit in retrospective if needed
----
-## Approval Gates
-Clear gates prevent scope creep and ensure quality. Know when to proceed vs escalate.
-### Risk Thresholds
-| Risk Level | Criteria | Action |
-|------------|----------|--------|
-| **Low** | Single repo, existing patterns, < 5 pts | Proceed without approval |
-| **Medium** | Cross-repo, new patterns, 5-13 pts | PM approval required |
-| **High** | Architecture change, external deps, 13+ pts | PM + Architect approval |
-| **Critical** | Breaking changes, security, data migration | Full team review |
-### Who Approves What
-| Decision Type | Approver | Escalation Path |
-|---------------|----------|-----------------|
-| Story prioritization | PM | → Orchestrator |
-| Technical approach | Architect | → PM → Orchestrator |
-| Sprint commitment | SM + Team | → PM |
-| Pattern deviation | Architect | → Orchestrator |
-| Process change | Orchestrator | → Team consensus |
-### Escalation Protocol
-When blocked or uncertain:
-```
-1. Document the blocker in session file
-2. Identify the appropriate approver (table above)
-3. Present options with trade-offs, not just problems
-4. Request decision with deadline
-5. If no response: escalate to next level
-6. Document final decision in sidecar
-```
-### Proceed vs Escalate Decision Tree
-```
-Is this within my role's authority?
-    ├─ YES → Does it follow existing patterns?
-    │           ├─ YES → Proceed
-    │           └─ NO → Escalate to Architect
-    └─ NO → Escalate to appropriate approver
-```
----
-## Sprint Planning Ceremony
-Structured workflow for sprint planning. PM leads, SM facilitates, Architect advises.
-### Pre-Planning (PM, async)
-1. **Backlog Grooming:**
-   - Review backlog.yaml for candidate stories
-   - Ensure stories have clear acceptance criteria
-   - Flag stories needing architecture review
-2. **Capacity Check:**
-   - Note team velocity from previous sprints
-   - Identify any planned absences or blockers
-   - Set preliminary sprint goal
-3. **Priority Stack:**
-   - Rank stories by business value
-   - Consider dependencies between stories
-   - Prepare recommendation for planning session
-### Planning Session (PM + SM + Architect)
-**Duration:** ~30 minutes per 2-week sprint
-1. **Sprint Goal (5 min):**
-   - PM proposes sprint goal
-   - Team discusses and refines
-   - Goal captured in `sprint/current-sprint.yaml`
-2. **Story Review (15 min):**
-   - PM presents prioritized stories
-   - Architect flags technical concerns
-   - SM notes dependencies and blockers
-3. **Sizing (5 min):**
-   - Team confirms or adjusts story points
-   - Stories > 8 pts flagged for breakdown
-   - Architect provides complexity input
-4. **Commitment (5 min):**
-   - Team commits to sprint scope
-   - SM updates sprint YAML with committed stories
-   - Any stretch goals identified
-### Post-Planning (SM)
-```bash
-# Update sprint file
-vim sprint/current-sprint.yaml
-# Create epic context if new epic
-/start-epic {epic-id}
-# Sprint is ready for /new-work
-```
-### Sprint Planning Checklist
-- [ ] Sprint goal defined and agreed
-- [ ] Stories prioritized and sized
-- [ ] Dependencies identified
-- [ ] Architect reviewed technical risk
-- [ ] Team committed to scope
-- [ ] Sprint YAML updated
-- [ ] Epic context generated (if new epic)
----
-## Handoffs to Tactical Agents
-Strategic agents hand off to tactical agents for implementation:
-| From | To | Trigger |
-|------|----|---------|
-| SM | TEA | Story ready for tests |
-| Architect | Dev | Design complete, ready to implement |
-| DevOps | Dev | Infrastructure ready for code |
-### Handoff Protocol
-1. Write context to session file
-2. Capture any learnings to sidecar
-3. Announce handoff with clear next steps
----
-## Decision Documentation
-Strategic agents make decisions that affect future work. Document them:
-```markdown
-# In decisions.md sidecar:
----
-## 2025-01-15 Sprint Planning: API-first approach
-**Context:** Deciding whether to build UI or API first for auth feature
-**Decision:** API-first - allows parallel UI development once contracts defined
-**Rationale:** Reduces blocking, enables contract testing
-**Affects:** All auth-related stories in Sprint 23
-```
----
-## Exit Protocol
-Before exiting:
-1. **Load shared behavior** - capture learnings to sidecar
-2. **Update sprint status** if changed
-3. **Document any pending decisions**
-4. **Clear handoff** if work continues with another agent
----
-**Strategic agents see the whole. They plan, coordinate, and delegate.**