mindforge-cc 11.9.0 → 11.9.2

package/.agent/mindforge/consult.md

@@ -31,7 +31,7 @@ $ARGUMENTS
    - Identify divergence (models disagree — flag for user)
    - Identify novel insights (unique points from one model)
 7. **Present results** with clear attribution (which model said what).
-8. Log consultation in token-ledger.jsonl (model, tokens, cost per call).
+8. Log consultation in token-usage.jsonl (model, tokens, cost per call).
 9. Report total cost of this consultation.
 10. Remind: "External opinions are advisory — user sovereignty applies."
 </process>

package/.agent/mindforge/cost-report.md

@@ -19,7 +19,7 @@ $ARGUMENTS
 
 <process>
 1. Parse flags: --window (time period, default: current session), --by-model, --by-task, --phase.
-2. Read `.mindforge/metrics/token-ledger.jsonl` for the specified window.
+2. Read `.mindforge/metrics/token-usage.jsonl` for the specified window.
 3. **Compute totals:**
    - Total tokens (input, output, cached)
    - Total estimated cost (USD)

package/.agent/mindforge/ship.md

@@ -29,7 +29,7 @@ Generate a CHANGELOG.md entry following Keep a Changelog format:
 - [Security improvements]
 ```
 
-Prepend this to CHANGELOG.md.
+Create changelogs/vX.Y.Z.md with the full entry. Prepend a matching entry to root CHANGELOG.md (trim oldest if rolling window > 10 entries).
 
 ## Step 2 — Run final quality gates
 Run all of the following and report results:
@@ -93,7 +93,7 @@ Generate a complete PR description:
 
 ## Step 4 — Commit and tag
 ```bash
-git add CHANGELOG.md
+git add changelogs/vX.Y.Z.md CHANGELOG.md
 git commit -m "docs(changelog): add Phase [N] release notes"
 git push origin feat/mindforge-core-scaffold
 ```

package/.agent/mindforge/update.md

@@ -15,7 +15,7 @@ Execute `bin/updater/self-update.js` `checkAndUpdate()`.
 Always show: current version, latest version, update type (major/minor/patch).
 
 ### 2. Changelog display (unless --skip-changelog)
-Fetch and display the relevant CHANGELOG.md section.
+Fetch and display the relevant changelog section (root CHANGELOG.md rolling window covers the newest 10 versions; older versions are fetched from changelogs/vX.Y.Z.md via the changelog-fetcher.js fallback path).
 For major updates: prefix with ⚠️ BREAKING CHANGES notice.
 Limit display to 3,000 characters — link to full CHANGELOG for the rest.
 

package/.agent/mindforge/wf-catalog.md

@@ -1,9 +1,9 @@
 ---
-description: "Browse the MindForge dynamic workflow library — 32 pre-built multi-agent workflows across 5 tiers"
+description: "Browse the MindForge dynamic workflow library — 35 pre-built multi-agent workflows across 5 tiers"
 ---
 # /mindforge:wf-catalog
 
-Browse and discover the MindForge **Dynamic Workflow Library** — 32 pre-built multi-agent workflows that run via Claude Code's Workflow tool.
+Browse and discover the MindForge **Dynamic Workflow Library** — 35 pre-built multi-agent workflows that run via Claude Code's Workflow tool.
 
 ## Usage
 `/mindforge:wf-catalog` — show all workflows grouped by tier
@@ -30,6 +30,8 @@ Browse and discover the MindForge **Dynamic Workflow Library** — 32 pre-built
 - `/mindforge:wf-mutation-testing` — Mutant generator → parallel kill-test → score
 - `/mindforge:wf-code-explainer` — Structure → domain → architecture → narrative tour
 - `/mindforge:wf-design-system-audit` — 5-dimension parallel audit → consistency score
+- `/mindforge:wf-orchestrate-review` — 4-lens panel (factual/domain/safety/style) → merged verdict
+- `/mindforge:wf-verification-loop` — Implementer vs. scorer loop → re-attempts until score maxed
 
 **Ops** — Infrastructure and release workflows
 - `/mindforge:wf-incident-response` — Parallel investigation → RCA → postmortem
@@ -46,6 +48,7 @@ Browse and discover the MindForge **Dynamic Workflow Library** — 32 pre-built
 - `/mindforge:wf-documentation-gen` — Parallel doc gen → normalize → publish-ready
 - `/mindforge:wf-api-migration` — Breaking change detection → guide → compat matrix
 - `/mindforge:wf-data-pipeline-validate` — Stage-by-stage validation → quality gates
+- `/mindforge:wf-workflow-optimizer` — Builder + scorer + optimizer → revises the process itself
 
 **Beast** — Compound workflows (5 phases, 8+ agents, adversarial verification)
 - `/mindforge:wf-security-hardening` — 5-angle OWASP scout → 3-vote verify → STRIDE → roadmap

package/.claude/commands/mindforge/consult.md

@@ -31,7 +31,7 @@ $ARGUMENTS
    - Identify divergence (models disagree — flag for user)
    - Identify novel insights (unique points from one model)
 7. **Present results** with clear attribution (which model said what).
-8. Log consultation in token-ledger.jsonl (model, tokens, cost per call).
+8. Log consultation in token-usage.jsonl (model, tokens, cost per call).
 9. Report total cost of this consultation.
 10. Remind: "External opinions are advisory — user sovereignty applies."
 </process>

package/.claude/commands/mindforge/cost-report.md

@@ -19,7 +19,7 @@ $ARGUMENTS
 
 <process>
 1. Parse flags: --window (time period, default: current session), --by-model, --by-task, --phase.
-2. Read `.mindforge/metrics/token-ledger.jsonl` for the specified window.
+2. Read `.mindforge/metrics/token-usage.jsonl` for the specified window.
 3. **Compute totals:**
    - Total tokens (input, output, cached)
    - Total estimated cost (USD)

package/.claude/commands/mindforge/ship.md

@@ -29,7 +29,7 @@ Generate a CHANGELOG.md entry following Keep a Changelog format:
 - [Security improvements]
 ```
 
-Prepend this to CHANGELOG.md.
+Create changelogs/vX.Y.Z.md with the full entry. Prepend a matching entry to root CHANGELOG.md (trim oldest if rolling window > 10 entries).
 
 ## Step 2 — Run final quality gates
 Run all of the following and report results:
@@ -93,7 +93,7 @@ Generate a complete PR description:
 
 ## Step 4 — Commit and tag
 ```bash
-git add CHANGELOG.md
+git add changelogs/vX.Y.Z.md CHANGELOG.md
 git commit -m "docs(changelog): add Phase [N] release notes"
 git push origin feat/mindforge-core-scaffold
 ```

package/.claude/commands/mindforge/skill-tdd.md

@@ -0,0 +1,53 @@
+---
+description: "Test-Driven Development — RED-GREEN-REFACTOR discipline. Write the failing test first, always."
+---
+
+# MindForge — TDD Skill Command
+# Usage: /mindforge:skill-tdd [feature or bug to implement]
+
+Activates the test-driven development skill. The core rule: **if you didn't watch the test fail, you don't know if it tests the right thing.**
+
+## Activation
+
+Load `.mindforge/skills/test-driven-development/SKILL.md` immediately.
+Follow its RED-GREEN-REFACTOR cycle strictly for the full duration of this session.
+
+## The Cycle (non-negotiable)
+
+### RED — Write a failing test
+1. Identify the smallest next behavior to implement.
+2. Write one test that asserts that behavior.
+3. Run it — confirm it FAILS for the right reason (not a syntax error, not a missing import — the actual assertion fails).
+4. Do not proceed until the test fails correctly.
+
+### GREEN — Write minimal code to pass
+1. Write the simplest code that makes the test pass.
+2. No gold-plating, no extra features. Minimum viable.
+3. Run the test — it must pass.
+4. Run the full suite — no regressions.
+
+### REFACTOR — Clean up
+1. Improve code structure, naming, and clarity.
+2. Remove duplication.
+3. Tests stay green throughout — run after every change.
+
+## Mandatory gates
+
+- **Never write code without a failing test.** Not for "obvious" cases, not for "quick" fixes, not for "trivial" implementations.
+- **One cycle at a time.** Complete RED-GREEN-REFACTOR before starting the next behavior.
+- **A passing test suite is always the starting state.** If tests are red when you begin, fix them first.
+
+## When the user asks for a feature
+
+1. Decompose into the smallest testable behavior.
+2. Start with the first RED step before writing any implementation.
+3. Repeat the cycle for each behavior.
+
+## When the user asks for a bug fix
+
+1. Write a failing test that reproduces the bug (this is your RED step).
+2. Confirm it fails.
+3. Fix the code (GREEN).
+4. Refactor if needed.
+
+The bug test becomes the regression guard — it must remain in the suite permanently.

package/.claude/commands/mindforge/skills-index.md

@@ -0,0 +1,118 @@
+---
+description: "Browse all available skills by category — discover and activate any skill by name."
+---
+
+# MindForge — Skills Index
+# Usage: /mindforge:skills-index [optional: category or keyword filter]
+
+Lists all skills available in this MindForge installation. Skills in the **Engine** tier activate automatically via trigger-matching. Skills in the **Extended** tier require explicit invocation.
+
+---
+
+## How to activate a skill
+
+**Engine tier** (auto-triggers from `.mindforge/skills/`): Just describe the task — the skill-loader matches your request against each skill's `triggers:` field and loads the relevant skill automatically.
+
+**Extended tier** (explicit from `.agent/skills/`): Ask Claude to "use the [skill-name] skill" or invoke it by name.
+
+---
+
+## Engine Tier Skills (auto-triggered)
+
+### Software Development
+| Skill | Triggers |
+|---|---|
+| `systematic-debugging` | systematic debugging, root cause analysis, debug methodology, 4-phase debug |
+| `test-driven-development` | TDD, red green refactor, write test first, test before code |
+| `plan` | write a plan, plan mode, implementation plan, plan before coding |
+| `simplify-code` | simplify code, clean up code, refactor for clarity, reduce complexity |
+| `requesting-code-review` | request code review, code review protocol, review this PR |
+| `spike` | technical spike, time-boxed spike, explore this problem |
+| `subagent-driven-development` | subagent driven development, delegate to subagent, multi-agent implementation |
+| `code-wiki` | code wiki, document codebase, knowledge wiki, explain codebase |
+
+### DevOps & Orchestration
+| Skill | Triggers |
+|---|---|
+| `kanban-orchestrator` | kanban orchestrator, multi-agent kanban, decompose and route, orchestrate tasks |
+| `kanban-worker` | kanban worker, pick up kanban task, complete kanban card |
+
+### GitHub Workflows
+| Skill | Triggers |
+|---|---|
+| `github-code-review` | github code review, review PR, pull request review workflow |
+| `github-pr-workflow` | github pr workflow, pull request lifecycle, open PR, merge PR |
+| `github-issues` | github issues, create issue, manage issues, issue triage |
+| `codebase-inspection` | codebase inspection, explore codebase, understand repository |
+
+### Research & Intelligence
+| Skill | Triggers |
+|---|---|
+| `research-paper-writing` | research paper, academic paper, write paper, arxiv paper |
+| `arxiv` | arxiv search, find papers, search arxiv, academic literature |
+| `osint-investigation` | OSINT investigation, public records research, entity investigation |
+| `domain-intel` | domain intelligence, investigate domain, domain research |
+| `duckduckgo-search` | duckduckgo search, DDG search, web search |
+| `scrapling` | scrape website, web scraping, extract web content |
+| `blogwatcher` | monitor blog, watch blog, track blog updates |
+
+### Creative
+| Skill | Triggers |
+|---|---|
+| `concept-diagrams` | concept diagram, educational diagram, SVG diagram, visual explanation |
+| `creative-ideation` | creative ideation, brainstorm ideas, creative ideas, generate concepts |
+| `pixel-art` | pixel art, create pixel art, sprite design |
+| `meme-generation` | meme generation, create meme, generate meme |
+
+### Security
+| Skill | Triggers |
+|---|---|
+| `web-pentest` | web penetration test, pentest this app, security test web app, OWASP test |
+| `oss-forensics` | OSS forensics, open source forensics, supply chain audit |
+| `sherlock` | sherlock, username investigation, find accounts, OSINT username |
+
+### Data & Tooling
+| Skill | Triggers |
+|---|---|
+| `jupyter-live-kernel` | jupyter kernel, live jupyter, interactive notebook |
+| `obsidian` | obsidian notes, obsidian vault, obsidian workflow |
+
+---
+
+## Extended Tier Skills (`.agent/skills/`, explicit activation)
+
+### Software Development
+`node-inspect-debugger` · `python-debugpy` · `skill-authoring` · `rest-graphql-debug`
+
+### GitHub
+`github-auth` · `github-repo-management`
+
+### DevOps
+`docker-management` · `devops-cli` · `devops-watchers` · `pinggy-tunnel` · `s6-container-supervision`
+
+### Research
+`llm-wiki` · `polymarket` · `parallel-cli`
+
+### Security
+`godmode` · `1password-skill`
+
+### Creative
+`hyperframes` · `article-illustrator` · `comic-creator` · `video-orchestrator`
+
+---
+
+## Usage examples
+
+```
+"Debug this null pointer — use systematic debugging"
+→ Engine tier: systematic-debugging activates automatically
+
+"I want to do TDD on this new auth module"
+→ Engine tier: test-driven-development activates automatically
+
+"Use the docker-management skill to set up my containers"
+→ Extended tier: explicit invocation of docker-management
+
+"Run an OSINT investigation on this company"
+→ Engine tier: osint-investigation activates automatically
+```

package/.claude/commands/mindforge/systematic-debug.md

@@ -0,0 +1,60 @@
+---
+description: "4-phase root cause debugging — understand the bug completely before attempting any fix."
+---
+
+# MindForge — Systematic Debug Command
+# Usage: /mindforge:systematic-debug [problem description]
+
+Activates the systematic debugging skill. The iron law: **NO FIXES WITHOUT ROOT CAUSE INVESTIGATION FIRST**. Symptom fixes are failure.
+
+## Activation
+
+Load `.mindforge/skills/systematic-debugging/SKILL.md` immediately.
+Follow its 4-phase protocol for the full duration of this session.
+
+## Phase 1 — Understand the Bug
+
+Before touching any code:
+
+1. **Reproduce it.** Can you reproduce it in a minimal, isolated environment?
+   - If no: document what you tried. Ask the user for more context. Do not guess.
+   - If yes: record the exact reproduction steps.
+
+2. **Read the error completely.** Stack trace, logs, error message — read every line.
+
+3. **Identify the invariant.** What assumption is being violated? What should be true that isn't?
+
+4. **Map the control flow.** Trace the path from input to failure point.
+
+**Output of Phase 1:** A written statement of the root cause hypothesis with evidence.
+
+## Phase 2 — Isolate the Root Cause
+
+1. Write a failing test that exercises exactly the broken invariant.
+2. Confirm the test fails for the right reason (not just any reason).
+3. Narrow scope: is the bug in this file? This function? This line?
+4. Check: is this a regression? Run `git log --oneline -20` on affected files.
+
+**Gate:** Do not proceed to Phase 3 without a failing test that proves the bug.
+
+## Phase 3 — Fix
+
+1. Apply the minimal fix that restores the invariant.
+2. Do not fix adjacent issues or refactor — single responsibility per fix.
+3. Run the failing test — it must now pass.
+4. Run the full test suite — zero new failures allowed.
+
+## Phase 4 — Verify and Document
+
+1. Confirm the original reproduction steps no longer trigger the bug.
+2. Write a one-paragraph RCA summary: what was broken, why, how it was fixed.
+3. Commit with a message that explains the root cause, not the symptom.
+
+## When you cannot find the root cause
+
+- Add logging/instrumentation at the point of failure.
+- Form 2–3 competing hypotheses and test each independently.
+- Document what you ruled out — negative evidence is evidence.
+- Ask the user for additional context before guessing.
+
+Do not apply a fix that "might work." Every fix requires a root cause explanation.

package/.claude/commands/mindforge/update.md

@@ -15,7 +15,7 @@ Execute `bin/updater/self-update.js` `checkAndUpdate()`.
 Always show: current version, latest version, update type (major/minor/patch).
 
 ### 2. Changelog display (unless --skip-changelog)
-Fetch and display the relevant CHANGELOG.md section.
+Fetch and display the relevant changelog section (root CHANGELOG.md rolling window covers the newest 10 versions; older versions are fetched from changelogs/vX.Y.Z.md via the changelog-fetcher.js fallback path).
 For major updates: prefix with ⚠️ BREAKING CHANGES notice.
 Limit display to 3,000 characters — link to full CHANGELOG for the rest.
 

package/.claude/commands/mindforge/wf-accessibility-audit.md

@@ -0,0 +1,31 @@
+---
+description: "WCAG 2.2 parallel per-criterion audit → 3-vote adversarial verify failures → remediation spec"
+---
+# /mindforge:wf-accessibility-audit
+
+Runs the **Accessibility Audit** dynamic workflow.
+
+## Usage
+`/mindforge:wf-accessibility-audit <UI codebase path or component>`
+
+## What it does
+- **Scope**: Define target UI and map components/pages to audit
+- **Audit**: 6 parallel WCAG principle auditors (Perceivable/Operable/Understandable/Robust + ARIA + Keyboard)
+- **Verify**: 3-vote adversarial verification of all Level A and AA failures
+- **Spec**: Remediation spec with exact ARIA attributes, HTML fixes, and WCAG references
+
+## Running
+
+Invoke via Claude Code's Workflow tool:
+
+```
+Workflow({
+  scriptPath: ".mindforge/dynamic-workflows/scripts/accessibility-audit.js",
+  args: "<your input>"
+})
+```
+
+Or discover via CLI:
+```bash
+node bin/mindforge-cli.js workflow info accessibility-audit
+```

package/.claude/commands/mindforge/wf-ai-model-eval.md

@@ -0,0 +1,31 @@
+---
+description: "4-parallel model benchmark agents → scoring matrix → cost/performance recommendation"
+---
+# /mindforge:wf-ai-model-eval
+
+Runs the **AI Model Evaluation** dynamic workflow.
+
+## Usage
+`/mindforge:wf-ai-model-eval <use case description>`
+
+## What it does
+- **Scope**: Define evaluation criteria and test prompts for the use case
+- **Benchmark**: 4 parallel model evaluators (quality / reasoning / speed / cost)
+- **Score**: Scoring matrix across all dimensions
+- **Recommend**: Ranked recommendation with break-even cost analysis
+
+## Running
+
+Invoke via Claude Code's Workflow tool:
+
+```
+Workflow({
+  scriptPath: ".mindforge/dynamic-workflows/scripts/ai-model-eval.js",
+  args: "<your input>"
+})
+```
+
+Or discover via CLI:
+```bash
+node bin/mindforge-cli.js workflow info ai-model-eval
+```

package/.claude/commands/mindforge/wf-api-migration.md

@@ -0,0 +1,31 @@
+---
+description: "Breaking change detection → versioning strategy → migration guide → compatibility matrix"
+---
+# /mindforge:wf-api-migration
+
+Runs the **API Migration** dynamic workflow.
+
+## Usage
+`/mindforge:wf-api-migration <old API path or description of changes>`
+
+## What it does
+- **Detect**: Detect breaking vs non-breaking changes between old and new API versions
+- **Version**: Propose versioning strategy: semver, URL versioning, or header versioning
+- **Guide**: Generate migration guide for API consumers
+- **Matrix**: Compatibility matrix: which client versions work with which API versions
+
+## Running
+
+Invoke via Claude Code's Workflow tool:
+
+```
+Workflow({
+  scriptPath: ".mindforge/dynamic-workflows/scripts/api-migration.js",
+  args: "<your input>"
+})
+```
+
+Or discover via CLI:
+```bash
+node bin/mindforge-cli.js workflow info api-migration
+```

package/.claude/commands/mindforge/wf-architecture-modernization.md

@@ -0,0 +1,32 @@
+---
+description: "Legacy architecture map → target design → migration sequencing → risk gates"
+---
+# /mindforge:wf-architecture-modernization
+
+Runs the **Architecture Modernization** dynamic workflow.
+
+## Usage
+`/mindforge:wf-architecture-modernization <codebase or modernization goal>`
+
+## What it does
+- **Map**: Map current architecture: components, dependencies, coupling, pain points
+- **Design**: 3 parallel target architecture proposals with trade-off analysis
+- **Select**: Judge panel selects best design, synthesizes hybrid
+- **Sequence**: Migration sequencing with risk gates and rollback checkpoints
+- **Roadmap**: Sprint-by-sprint modernization roadmap
+
+## Running
+
+Invoke via Claude Code's Workflow tool:
+
+```
+Workflow({
+  scriptPath: ".mindforge/dynamic-workflows/scripts/architecture-modernization.js",
+  args: "<your input>"
+})
+```
+
+Or discover via CLI:
+```bash
+node bin/mindforge-cli.js workflow info architecture-modernization
+```

package/.claude/commands/mindforge/wf-catalog.md

@@ -0,0 +1,62 @@
+---
+description: "Browse the MindForge dynamic workflow library — 35 pre-built multi-agent workflows across 5 tiers"
+---
+# /mindforge:wf-catalog
+
+Browse and discover the MindForge **Dynamic Workflow Library** — 35 pre-built multi-agent workflows that run via Claude Code's Workflow tool.
+
+## Usage
+`/mindforge:wf-catalog` — show all workflows grouped by tier
+
+## Tiers
+
+**Research** — Fan-out search, adversarial verification, cited synthesis
+- `/mindforge:wf-competitive-analysis` — SWOT and positioning summary
+- `/mindforge:wf-tech-evaluation` — Scored technology comparison matrix
+- `/mindforge:wf-ai-model-eval` — Benchmark AI models → scoring matrix → recommendation
+- `/mindforge:wf-ux-heuristic-audit` — 10 Nielsen heuristics parallel audit → fix brief
+- `/mindforge:wf-competitive-teardown` — 5 competitor angles → positioning report
+
+**Dev** — Coding-assistant power workflows
+- `/mindforge:wf-code-audit` — Parallel security + quality + performance audit
+- `/mindforge:wf-feature-planner` — Brief → PRD → architecture → user stories
+- `/mindforge:wf-pr-review` — 4-dimensional parallel PR review
+- `/mindforge:wf-tdd-sprint` — Strict Red-Green-Refactor TDD loop
+- `/mindforge:wf-refactor-plan` — Technical debt scan → safe refactor plan
+- `/mindforge:wf-test-coverage-gap` — Per-module coverage analysis → test-writing plan
+- `/mindforge:wf-api-contract-test` — Spec vs impl Writer/Reviewer → violation report
+- `/mindforge:wf-debug-detective` — 4-hypothesis parallel investigation → scientific RCA
+- `/mindforge:wf-writer-reviewer` — Implement → fresh context review → verdict
+- `/mindforge:wf-mutation-testing` — Mutant generator → parallel kill-test → score
+- `/mindforge:wf-code-explainer` — Structure → domain → architecture → narrative tour
+- `/mindforge:wf-design-system-audit` — 5-dimension parallel audit → consistency score
+- `/mindforge:wf-orchestrate-review` — 4-lens panel (factual/domain/safety/style) → merged verdict
+- `/mindforge:wf-verification-loop` — Implementer vs. scorer loop → re-attempts until score maxed
+
+**Ops** — Infrastructure and release workflows
+- `/mindforge:wf-incident-response` — Parallel investigation → RCA → postmortem
+- `/mindforge:wf-release-prep` — Tests → changelog → version bump → PR
+- `/mindforge:wf-dependency-health` — CVE + license + staleness audit → risk matrix
+- `/mindforge:wf-database-migration` — Schema diff → risk → scripts → runbook
+- `/mindforge:wf-multi-repo-sync` — Per-repo audit → divergence map → sync plan
+- `/mindforge:wf-cost-analysis` — Infra/API/query/bundle cost agents → ROI plan
+
+**Intelligence** — Deep analysis and optimization
+- `/mindforge:wf-onboard-codebase` — Map → domain → architecture → guided tour
+- `/mindforge:wf-perf-optimize` — Profile → bottleneck hunt → prioritized fix plan
+- `/mindforge:wf-architecture-modernization` — Legacy map → 3 designs → migration roadmap
+- `/mindforge:wf-documentation-gen` — Parallel doc gen → normalize → publish-ready
+- `/mindforge:wf-api-migration` — Breaking change detection → guide → compat matrix
+- `/mindforge:wf-data-pipeline-validate` — Stage-by-stage validation → quality gates
+- `/mindforge:wf-workflow-optimizer` — Builder + scorer + optimizer → revises the process itself
+
+**Beast** — Compound workflows (5 phases, 8+ agents, adversarial verification)
+- `/mindforge:wf-security-hardening` — 5-angle OWASP scout → 3-vote verify → STRIDE → roadmap
+- `/mindforge:wf-accessibility-audit` — WCAG 2.2 per-criterion → 3-vote verify → remediation spec
+- `/mindforge:wf-security-threat-model` — Asset inventory → STRIDE x6 → mitigations → CVSS matrix
+
+## CLI Discovery
+```bash
+node bin/mindforge-cli.js workflow list
+node bin/mindforge-cli.js workflow info <name>
+```

package/.claude/commands/mindforge/wf-code-audit.md

@@ -0,0 +1,31 @@
+---
+description: "Parallel security + quality + performance audit with adversarial finding verification"
+---
+# /mindforge:wf-code-audit
+
+Runs the **Code Audit** dynamic workflow.
+
+## Usage
+`/mindforge:wf-code-audit [path or 'current git diff']`
+
+## What it does
+- **Scope**: Builds file list from git diff, specified path, or entire codebase
+- **Audit**: 3 parallel auditors — OWASP security, code quality, performance patterns
+- **Verify**: Adversarial 2-vote verification for all high/critical severity findings
+- **Report**: Risk-ranked report with remediation steps for each confirmed finding
+
+## Running
+
+Invoke via Claude Code's Workflow tool:
+
+```
+Workflow({
+  scriptPath: ".mindforge/dynamic-workflows/scripts/code-audit.js",
+  args: "<your input>"
+})
+```
+
+Or discover via CLI:
+```bash
+node bin/mindforge-cli.js workflow info code-audit
+```

package/.claude/commands/mindforge/wf-code-explainer.md

@@ -0,0 +1,31 @@
+---
+description: "Structural map → domain extraction → architecture patterns → guided narrative tour for onboarding"
+---
+# /mindforge:wf-code-explainer
+
+Runs the **Code Explainer** dynamic workflow.
+
+## Usage
+`/mindforge:wf-code-explainer <codebase path or module to explain>`
+
+## What it does
+- **Structure**: Map file structure, entry points, and module boundaries
+- **Domain**: Extract domain concepts, business logic, and key abstractions
+- **Architecture**: Identify architectural patterns, data flow, and design decisions
+- **Tour**: Synthesize a guided narrative tour for a new developer
+
+## Running
+
+Invoke via Claude Code's Workflow tool:
+
+```
+Workflow({
+  scriptPath: ".mindforge/dynamic-workflows/scripts/code-explainer.js",
+  args: "<your input>"
+})
+```
+
+Or discover via CLI:
+```bash
+node bin/mindforge-cli.js workflow info code-explainer
+```

package/.claude/commands/mindforge/wf-competitive-analysis.md

@@ -0,0 +1,31 @@
+---
+description: "Multi-angle competitive research producing a SWOT and positioning summary"
+---
+# /mindforge:wf-competitive-analysis
+
+Runs the **Competitive Analysis** dynamic workflow.
+
+## Usage
+`/mindforge:wf-competitive-analysis <product/company/technology>`
+
+## What it does
+- **Scope**: Defines the competitive landscape target
+- **Research**: 5 parallel angles — product features, pricing, reviews, community, roadmap
+- **SWOT**: Synthesizes strengths, weaknesses, opportunities, threats
+- **Position**: Produces strategic positioning and differentiation recommendations
+
+## Running
+
+Invoke via Claude Code's Workflow tool:
+
+```
+Workflow({
+  scriptPath: ".mindforge/dynamic-workflows/scripts/competitive-analysis.js",
+  args: "<your input>"
+})
+```
+
+Or discover via CLI:
+```bash
+node bin/mindforge-cli.js workflow info competitive-analysis
+```