@soleri/forge 9.0.0 → 9.2.0

package/src/skills/context-resume/SKILL.md

@@ -0,0 +1,95 @@
+---
+name: context-resume
+description: >
+  Use when the user says "where did I leave off", "what was I working on", "catch me up",
+  "resume", "continue where we stopped", or is starting a new session and needs to reconstruct
+  working context from memory, plans, and sessions.
+---
+
+# Context Resume — Pick Up Where You Left Off
+
+Reconstruct full working context in seconds. Chains memory, plans, sessions, and brain to rebuild exactly where you left off — even across session boundaries and context compactions.
+
+## Steps
+
+### 1. Load Active Plans
+
+```
+YOUR_AGENT_core op:plan_stats
+YOUR_AGENT_core op:get_plan
+YOUR_AGENT_core op:plan_list_tasks
+  params: { planId: "<id>" }
+```
+
+Present: plan objective, task status (completed/in-progress/pending), what's next.
+
+### 2. Search Recent Memory
+
+```
+YOUR_AGENT_core op:memory_search
+  params: { query: "session summary" }
+YOUR_AGENT_core op:memory_list
+YOUR_AGENT_core op:vault_recent
+```
+
+### 3. Check Active Loops
+
+```
+YOUR_AGENT_core op:loop_is_active
+YOUR_AGENT_core op:loop_status
+```
+
+### 4. Brain Snapshot
+
+```
+YOUR_AGENT_core op:brain_strengths
+```
+
+### 5. System Health
+
+```
+YOUR_AGENT_core op:admin_health
+```
+
+## Presenting the Resume
+
+```
+## Where You Left Off
+
+**Active Plans:**
+- [Plan name] — X/Y tasks complete, next: [task]
+
+**Last Session:**
+- [Summary — what was done, key decisions]
+
+**Recent Captures:**
+- [New patterns/anti-patterns added]
+
+**Active Loops:**
+- [Any in-progress validation loops]
+
+**Brain Says:**
+- [Top relevant patterns]
+
+**Health:** [OK / Issues]
+
+## Recommended Next Step
+[Based on active plans and last session context]
+```
+
+## Common Mistakes
+
+- Not checking for active loops (missing mid-flight TDD or debug cycles)
+- Skipping the health check (stale caches can cause confusing behavior)
+- Not loading recent vault captures (missing context from last session)
+
+## Quick Reference
+
+| Op | When to Use |
+|----|-------------|
+| `plan_stats` / `get_plan` / `plan_list_tasks` | Active plans |
+| `memory_search` / `memory_list` | Session summaries |
+| `vault_recent` | Recently captured knowledge |
+| `loop_is_active` / `loop_status` | In-flight loops |
+| `brain_strengths` | Relevant proven patterns |
+| `admin_health` | System health check |

package/src/skills/deep-review/SKILL.md

@@ -0,0 +1,186 @@
+---
+name: deep-review
+description: >
+  Use for in-depth code review beyond linting — architecture fitness, code smells, solution quality,
+  optimization opportunities. Triggers on "deep review", "review this code", "is this well architected",
+  "code smells", "review this module", "architecture review", "is this the right approach",
+  "optimization review". Works on any codebase. For vault-specific knowledge quality, use vault-smells instead.
+---
+
+# Deep Review — Architecture, Smells & Solution Quality
+
+Multi-pass code review that goes beyond surface lint. Analyzes structural health, code smells, architectural fitness, and solution quality. Works on any codebase — vault context is optional enrichment, not a requirement.
+
+## Input
+
+The user provides a **target**: file, module, directory, PR diff, or function. If unclear, ask.
+
+## The Three Passes
+
+### Pass 1: Structural Analysis & Code Smells
+
+**Metrics** (gather by reading the code):
+- File length and function count
+- Cyclomatic complexity (nesting depth, branch count)
+- Dependency count — imports from how many modules?
+- Export surface area — how much is public vs. should be internal?
+
+**Structural Smells:**
+- **God file/class** — too many responsibilities, >300 lines with mixed concerns
+- **Long parameter lists** — function takes 5+ params (should be an object/config)
+- **Deep nesting** — 4+ levels of if/for/try/catch
+- **Shotgun surgery** — changing this code requires touching 5+ other files
+- **Primitive obsession** — passing raw strings/numbers instead of domain types
+- **Boolean blindness** — functions with multiple boolean params (`fn(true, false, true)`)
+
+**Duplication Smells:**
+- Copy-paste with slight variations
+- Repeated conditional logic — same if-chain in 3+ places
+- Parallel structures that always change together
+
+**Temporal Smells** (check git history):
+- Files that always change together but live in different modules → missing abstraction
+- Functions that get patched repeatedly → wrong abstraction
+- High churn files → instability signal
+
+```
+git log --format=format: --name-only --since="3 months ago" <target-path> | sort | uniq -c | sort -rn | head -20
+```
+
+Present findings before moving to Pass 2.
+
+### Pass 2: Architectural Fitness
+
+**Dependency Direction:**
+- Do dependencies flow in the right direction? (Outer layers depend on inner, not reverse)
+- Are there circular dependencies?
+- Does the code reach across module boundaries it shouldn't?
+
+**Abstraction Level:**
+- Is this the right level of abstraction for the problem?
+- Over-engineered? (abstraction for one use case, premature generalization)
+- Under-engineered? (inline logic that should be extracted)
+
+**Cohesion & Coupling:**
+- Does everything in this module belong together? (high cohesion)
+- Is the module tangled with others? (low coupling desired)
+- Feature envy — does a function touch another module's internals more than its own?
+
+**Vault Context** (optional — only if vault is connected):
+
+```
+YOUR_AGENT_core op:search_intelligent
+  params: { query: "<module name> architecture pattern" }
+```
+
+If vault has relevant patterns, check alignment. If not, skip — this pass works without vault.
+
+### Pass 3: Solution Quality Assessment
+
+**Simplification:**
+- Is there a simpler way to achieve the same result?
+- Could any abstraction be removed without loss?
+- Are there standard library/framework features that replace custom code?
+
+**Edge Cases:**
+- What inputs would break this?
+- Are error paths handled or just the happy path?
+- What happens with empty/null/undefined inputs?
+- Concurrency: race conditions, shared mutable state?
+
+**Performance:**
+- Any O(n²) or worse hidden in loops?
+- Unnecessary allocations, copies, or serialization?
+- N+1 query patterns?
+- Unbounded growth (arrays/maps that grow without limit)?
+
+**Evolutionary Fitness:**
+- How does this code age? Easy to modify in 6 months?
+- Does it create "gravity" — attracting more complexity over time?
+- Are extension points in the right places?
+
+## Presenting the Report
+
+```
+## Deep Review: [target name]
+
+### Structural Health
+| Metric | Value | Verdict |
+|--------|-------|---------|
+| Lines | X | OK / ⚠️ Large |
+| Functions | X | OK / ⚠️ Many |
+| Max nesting | X | OK / ⚠️ Deep |
+| Dependencies | X | OK / ⚠️ Heavy |
+| Export surface | X public / Y total | OK / ⚠️ Wide |
+
+### Code Smells
+| Smell | Location | Severity | Detail |
+|-------|----------|----------|--------|
+| God file | file.ts | ⚠️ | 450 lines, 3 mixed concerns |
+| Feature envy | fn() → other module | ⚠️ | Reaches into X internals |
+| Deep nesting | line 120-180 | 💡 | 5 levels, consider early returns |
+
+### Architecture
+| Aspect | Assessment |
+|--------|------------|
+| Dependency direction | ✅ Clean / ⚠️ Reverse dep on X |
+| Abstraction level | ✅ Right / ⚠️ Over/Under |
+| Cohesion | ✅ High / ⚠️ Mixed concerns |
+| Coupling | ✅ Low / ⚠️ Tight with X |
+
+### Solution Quality
+| Area | Finding |
+|------|---------|
+| Simplification | [opportunity or "none found"] |
+| Edge cases | [gaps found or "well covered"] |
+| Performance | [concerns or "no issues"] |
+| Evolutionary fitness | [assessment] |
+
+### Recommendations
+| Priority | Action | Impact |
+|----------|--------|--------|
+| 1 | [most impactful] | High |
+| 2 | [second] | Medium |
+```
+
+## Severity Scale
+
+| Level | Meaning |
+|-------|---------|
+| ✅ | Clean — no action needed |
+| 💡 | Info — worth knowing, low priority |
+| ⚠️ | Warning — should fix, causes friction |
+| 🔴 | Critical — fix before shipping, causes bugs or blocks scaling |
+
+## Capturing Learnings (Optional)
+
+If the review uncovers a pattern or anti-pattern worth remembering:
+
+```
+YOUR_AGENT_core op:capture_knowledge
+  params: {
+    title: "<pattern name>",
+    description: "<what was found, why it matters>",
+    type: "pattern" | "anti-pattern",
+    category: "architecture",
+    tags: ["code-review", "<specific-tag>"]
+  }
+```
+
+Only capture if genuinely reusable — not every review finding is vault-worthy.
+
+## Common Mistakes
+
+- Reviewing without reading the full file first (missing context)
+- Reporting every minor style issue as a "smell" (noise kills signal)
+- Suggesting rewrites when the code is adequate (perfect is the enemy of good)
+- Skipping git history (temporal smells are the most actionable)
+- Treating all smells as equal severity (prioritize by impact)
+
+## Quick Reference
+
+| Pass | Focus | Key Activities |
+|------|-------|----------------|
+| 1. Structural | Metrics + Smells | Read code, check complexity, find smells, check git history |
+| 2. Architecture | Fitness | Dependency direction, abstraction level, cohesion/coupling |
+| 3. Solution | Quality | Simplification, edge cases, performance, evolution |

package/src/skills/executing-plans/SKILL.md

@@ -0,0 +1,91 @@
+---
+name: executing-plans
+description: >
+  Use when the user says "execute my plan", "run the plan", "start executing", "implement the
+  plan step by step", or has a written plan to execute sequentially with review checkpoints.
+  Tasks run one at a time in order. If tasks are independent and can run in parallel, use
+  parallel-execute instead.
+---
+
+# Executing Plans
+
+Load plan, review critically, execute tasks in batches, report for review between batches.
+
+**Announce at start:** "I'm using the executing-plans skill to implement this plan."
+
+## The Process
+
+### Step 1: Load and Review Plan
+
+```
+YOUR_AGENT_core op:get_plan
+YOUR_AGENT_core op:plan_list_tasks
+  params: { planId: "<id>" }
+YOUR_AGENT_core op:plan_stats
+```
+
+If no tracked plan exists, read from `docs/plans/`. Review critically — raise concerns before starting.
+
+### Step 2: Start Execution Loop
+
+```
+YOUR_AGENT_core op:loop_start
+  params: { prompt: "<plan objective>", mode: "custom" }
+```
+
+### Step 3: Execute Batch (default: first 3 tasks)
+
+For each task:
+1. `op:update_task` — mark `in_progress`
+2. Follow each step exactly
+3. Run verifications as specified
+4. `op:update_task` — mark `completed`
+5. `op:loop_iterate` — track progress
+
+### Step 4: Report
+
+Show what was implemented, verification output, loop status. Say: "Ready for feedback."
+
+### Step 5: Continue
+
+Apply feedback, execute next batch, repeat until complete.
+
+### Step 6: Complete Development
+
+1. Run final verification (use verification-before-completion skill)
+2. `YOUR_AGENT_core op:loop_complete`
+3. `YOUR_AGENT_core op:plan_reconcile` — compare planned vs actual
+4. `YOUR_AGENT_core op:plan_complete_lifecycle` — extract knowledge, archive
+5. `YOUR_AGENT_core op:session_capture` — save session context
+
+Capture mid-execution learnings with `op:capture_quick` as they happen — don't wait until the end.
+
+## When to Stop
+
+- Hit a blocker (missing dependency, unclear instruction, repeated test failures)
+- Plan has critical gaps
+- Don't understand an instruction
+
+**Ask for clarification rather than guessing.**
+
+## Common Mistakes
+
+- Not reviewing the plan critically before starting
+- Skipping verifications to save time
+- Guessing through blockers instead of stopping to ask
+- Forgetting to reconcile the plan after execution (drift data improves future plans)
+- Starting implementation on main/master without explicit consent
+
+## Quick Reference
+
+| Op | When to Use |
+|----|-------------|
+| `get_plan` / `plan_list_tasks` / `plan_stats` | Load plan |
+| `update_task` | Mark task status |
+| `loop_start` / `loop_iterate` / `loop_complete` | Validation loop |
+| `plan_reconcile` | Post-execution drift report |
+| `plan_complete_lifecycle` | Extract knowledge, archive |
+| `session_capture` | Save session context |
+| `capture_quick` | Mid-execution learnings |
+
+**Related skills:** writing-plans, verification-before-completion, test-driven-development

package/src/skills/fix-and-learn/SKILL.md

@@ -0,0 +1,102 @@
+---
+name: fix-and-learn
+description: >
+  Use AFTER a root cause has been identified (typically via systematic-debugging) to execute
+  the fix and capture the learning in the vault. Triggers on "fix it", "apply the fix",
+  "patch this and remember". Do NOT use as the first response to a bug — use
+  systematic-debugging first to find root cause.
+---
+
+# Fix & Learn — Debug, Repair, Capture
+
+Fix bugs through a structured recovery workflow, then capture the root cause as a reusable anti-pattern. The learning step makes fixes compound across sessions.
+
+## The Search Order — MANDATORY
+
+**Never jump to writing code.** Always follow this lookup order:
+
+1. **Vault first** — has this been solved before?
+2. **Web search** — is there a known solution?
+3. **Plan the fix** — design approach before touching code
+4. **Implement** — only after steps 1-3
+
+## Orchestration Sequence
+
+### Step 1: Classify and Route
+
+```
+YOUR_AGENT_core op:route_intent
+  params: { prompt: "<bug description>" }
+```
+
+### Step 2: Check Vault First
+
+```
+YOUR_AGENT_core op:search_intelligent
+  params: { query: "<error message or bug description>" }
+YOUR_AGENT_core op:memory_search
+  params: { query: "<bug description>" }
+```
+
+If vault returns a high-confidence match — use it. Don't re-investigate solved problems.
+
+### Step 3: Search the Web
+
+If vault has no answer, search for known issues, Stack Overflow answers, GitHub issues, official docs.
+
+### Step 4: Start Fix Loop
+
+```
+YOUR_AGENT_core op:loop_start
+  params: { prompt: "Fix: <bug description>", mode: "custom" }
+```
+
+### Step 5: Diagnose and Fix
+
+If Steps 2-3 didn't produce a solution, use systematic-debugging skill:
+
+1. Reproduce the issue
+2. Isolate root cause
+3. Plan the fix before writing code
+4. Implement the fix
+5. Verify — no regressions
+
+### Step 6: Validate
+
+Run test suite. Use verification-before-completion skill. Complete loop: `YOUR_AGENT_core op:loop_complete`.
+
+### Step 7: Capture the Learning
+
+```
+YOUR_AGENT_core op:capture_knowledge
+  params: {
+    title: "<bug title>",
+    description: "<root cause, solution, what made it hard to find>",
+    type: "anti-pattern",
+    category: "<domain>",
+    tags: ["<error-type>", "<component>"]
+  }
+```
+
+Run `YOUR_AGENT_core op:curator_detect_duplicates` to avoid redundant entries.
+
+## Exit Criteria
+
+Bug resolved, tests pass, root cause captured in vault. A fix without a capture is incomplete.
+
+## Common Mistakes
+
+- Jumping to code before searching vault/web
+- Skipping the capture step after fixing
+- Not running the full test suite to check regressions
+
+## Quick Reference
+
+| Op | When to Use |
+|----|-------------|
+| `route_intent` | Classify as FIX intent |
+| `search_intelligent` | Check vault for known bugs |
+| `memory_search` | Search session memories |
+| `loop_start` / `loop_iterate` / `loop_complete` | Iterative fix cycle |
+| `capture_knowledge` / `capture_quick` | Persist anti-pattern |
+| `curator_detect_duplicates` | Prevent redundant entries |

package/src/skills/health-check/SKILL.md

@@ -0,0 +1,135 @@
+---
+name: health-check
+description: >
+  Use when the user asks "check health", "system status", "how healthy is the vault",
+  "knowledge quality", "run diagnostics", "vault health report", or wants a read-only health
+  assessment of the knowledge base — scoring, reporting, finding issues. Does NOT modify vault
+  entries. To actively clean, merge, or deduplicate, use vault-curate instead.
+---
+
+# Health Check — Knowledge Base Maintenance
+
+Comprehensive maintenance cycle on the knowledge base. Finds stale entries, duplicates, contradictions, and quality issues.
+
+## Steps
+
+### 1. System Health
+
+```
+YOUR_AGENT_core op:admin_health
+YOUR_AGENT_core op:admin_diagnostic
+```
+
+### 2. Vault Metrics
+
+```
+YOUR_AGENT_core op:admin_vault_size
+YOUR_AGENT_core op:admin_vault_analytics
+YOUR_AGENT_core op:vault_domains
+YOUR_AGENT_core op:vault_tags
+```
+
+### 3. Quality Audit
+
+```
+YOUR_AGENT_core op:curator_health_audit
+```
+
+### 4. Find Duplicates
+
+```
+YOUR_AGENT_core op:curator_detect_duplicates
+```
+
+### 5. Find Contradictions
+
+```
+YOUR_AGENT_core op:curator_contradictions
+YOUR_AGENT_core op:curator_resolve_contradiction
+  params: { contradictionId: "<id>" }
+```
+
+### 6. Find Stale Entries
+
+```
+YOUR_AGENT_core op:vault_age_report
+```
+
+Entries >30 days without updates: refresh, archive, or delete.
+
+### 7. Check Search Quality
+
+```
+YOUR_AGENT_core op:admin_search_insights
+```
+
+### 8. Memory Health
+
+```
+YOUR_AGENT_core op:memory_stats
+YOUR_AGENT_core op:memory_deduplicate
+```
+
+### 9. Governance Queue
+
+```
+YOUR_AGENT_core op:governance_proposals params: { action: "list" }
+YOUR_AGENT_core op:governance_expire
+```
+
+### 10. Fix Everything (Optional, with user approval)
+
+- `op:curator_groom_all` — normalize tags, fix metadata
+- `op:curator_consolidate` — deduplicate, normalize, quality-score
+- `op:memory_prune` — remove stale memories
+- `op:brain_build_intelligence` — rebuild with clean data
+- `op:admin_reset_cache` — clear caches
+
+## Presenting the Report
+
+```
+## Knowledge Health Report
+
+### System
+| Check | Status |
+|-------|--------|
+| Infrastructure | OK / Issues |
+
+### Vault Quality
+| Metric | Value | Status |
+|--------|-------|--------|
+| Total entries | X | — |
+| Quality score | X/100 | Good/Warning/Critical |
+
+### Issues Found
+| Issue | Count | Action |
+|-------|-------|--------|
+| Duplicates | X | Merge |
+| Contradictions | X | Resolve |
+| Stale entries (>30d) | X | Review |
+| Search misses | X | Fill gaps |
+
+### Recommended Actions
+1. [Most impactful fix]
+2. [Second most impactful]
+```
+
+## Common Mistakes
+
+- Running cleanup without presenting findings to user first
+- Skipping search insights (missing knowledge gaps)
+- Not rebuilding brain intelligence after major cleanup
+
+## Quick Reference
+
+| Op | When to Use |
+|----|-------------|
+| `admin_health` / `admin_diagnostic` | System health |
+| `admin_vault_analytics` / `admin_vault_size` | Vault metrics |
+| `curator_health_audit` | Quality score |
+| `curator_detect_duplicates` | Find duplicates |
+| `curator_contradictions` | Find conflicts |
+| `vault_age_report` | Stale entries |
+| `admin_search_insights` | Search miss analysis |
+| `curator_consolidate` | Full cleanup pipeline |
+| `brain_build_intelligence` | Rebuild after cleanup |

package/src/skills/knowledge-harvest/SKILL.md

@@ -0,0 +1,97 @@
+---
+name: knowledge-harvest
+description: >
+  Use to EXTRACT multiple patterns from a source — code, docs, PRs, articles. Triggers on
+  "learn from this", "harvest knowledge", "ingest this document", "extract patterns from".
+  The agent reads the source and identifies what to capture. For saving a single known item,
+  use vault-capture instead.
+---
+
+# Knowledge Harvest — Extract Patterns From Anything
+
+Point at code, docs, PRs, architecture decisions, or postmortems — the agent extracts every pattern, anti-pattern, decision, and principle, then captures them to the vault.
+
+## Steps
+
+### 1. Understand the Source
+
+Read target content and classify: `YOUR_AGENT_core op:route_intent params: { prompt: "Extract knowledge from: <source>" }`
+
+### 2. Check What's Already Known
+
+```
+YOUR_AGENT_core op:search_intelligent
+  params: { query: "<topic of source material>" }
+YOUR_AGENT_core op:vault_tags
+YOUR_AGENT_core op:vault_domains
+```
+
+Focus extraction on gaps — skip what vault already covers.
+
+### 3. Extract and Classify
+
+| Type | What to Look For |
+|------|-----------------|
+| **pattern** | Repeatable approaches that work |
+| **anti-pattern** | Known mistakes to avoid |
+| **decision** | Architectural choices with rationale |
+| **principle** | Guiding rules or heuristics |
+| **workflow** | Step-by-step procedures |
+
+For each: determine category, severity, and tags.
+
+### 4. Batch Capture
+
+```
+YOUR_AGENT_core op:capture_knowledge
+  params: {
+    title: "<clear, searchable name>",
+    description: "<what it is, when to apply, why it matters>",
+    type: "<pattern|anti-pattern|decision|principle|workflow>",
+    category: "<domain>",
+    tags: ["<tag1>", "<tag2>"],
+    example: "<code snippet or quote>",
+    why: "<reasoning>"
+  }
+```
+
+Present each capture as you go: `Captured: api-auth-jwt (pattern, critical)`
+
+### 5. Post-Harvest Quality
+
+- `op:curator_detect_duplicates` — find duplicates created during harvest
+- `op:curator_groom_all` — normalize tags, fix metadata
+- `op:curator_contradictions` — check for conflicts
+
+### 6. Verify and Report
+
+```
+## Harvest Complete
+
+Source: [name]
+Extracted: X entries (Y patterns, Z anti-patterns, W decisions)
+Duplicates: N (merged/skipped)
+Contradictions: N (flagged)
+Vault health: OK
+```
+
+Optionally promote universal patterns: `op:memory_promote_to_global`.
+
+## Common Mistakes
+
+- Not checking vault before extracting (creates duplicates)
+- Capturing too-granular entries instead of atomic, searchable ones
+- Skipping post-harvest quality checks (duplicates and contradictions accumulate)
+
+## Quick Reference
+
+| Op | When to Use |
+|----|-------------|
+| `search_intelligent` | Check existing knowledge |
+| `vault_tags` / `vault_domains` | See what's covered |
+| `capture_knowledge` | Capture each item |
+| `curator_detect_duplicates` | Post-harvest dedup |
+| `curator_groom_all` | Normalize entries |
+| `curator_contradictions` | Find conflicts |
+| `memory_promote_to_global` | Share cross-project |
+| `admin_vault_analytics` | Knowledge quality |