mindforge-cc 11.8.3 → 11.9.1

package/.agent/skills/subagent-driven-development/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: subagent-driven-development
-description: "Execute plans via delegate_task subagents (2-stage review)."
+description: "Execute plans via Agent subagents (2-stage review)."
 version: 1.1.0
 ---
 
@@ -34,7 +34,7 @@ Read the plan file. Extract ALL tasks with their full text and context upfront.
 
 ```python
 # Read the plan
-read_file("docs/plans/feature-plan.md")
+Read("docs/plans/feature-plan.md")
 
 # Create todo list with all tasks
 todo([
@@ -52,10 +52,10 @@ For EACH task in the plan:
 
 #### Step 1: Dispatch Implementer Subagent
 
-Use `delegate_task` with complete context:
+Use `Agent` with complete context:
 
 ```python
-delegate_task(
+Agent(
     goal="Implement Task 1: Create User model with email and password_hash fields",
     context="""
     TASK FROM PLAN:
@@ -78,7 +78,6 @@ delegate_task(
     - Tests use pytest, run from project root
     - bcrypt already in requirements.txt
     """,
-    toolsets=['terminal', 'file']
 )
 ```
 
@@ -87,7 +86,7 @@ delegate_task(
 After the implementer completes, verify against the original spec:
 
 ```python
-delegate_task(
+Agent(
     goal="Review if implementation matches the spec from the plan",
     context="""
     ORIGINAL TASK SPEC:
@@ -105,7 +104,6 @@ delegate_task(
 
     OUTPUT: PASS or list of specific spec gaps to fix.
     """,
-    toolsets=['file']
 )
 ```
 
@@ -116,7 +114,7 @@ delegate_task(
 After spec compliance passes:
 
 ```python
-delegate_task(
+Agent(
     goal="Review code quality for Task 1 implementation",
     context="""
     FILES TO REVIEW:
@@ -137,7 +135,6 @@ delegate_task(
     - Minor Issues: [optional]
     - Verdict: APPROVED or REQUEST_CHANGES
     """,
-    toolsets=['file']
 )
 ```
 
@@ -154,7 +151,7 @@ todo([{"id": "task-1", "content": "Create User model with email field", "status"
 After ALL tasks are complete, dispatch a final integration reviewer:
 
 ```python
-delegate_task(
+Agent(
     goal="Review the entire implementation for consistency and integration issues",
     context="""
     All tasks from the plan are complete. Review the full implementation:
@@ -163,7 +160,6 @@ delegate_task(
     - All tests passing?
     - Ready for merge?
     """,
-    toolsets=['terminal', 'file']
 )
 ```
 

package/.agent/skills/systematic-debugging/SKILL.md

@@ -61,7 +61,7 @@ You MUST complete each phase before proceeding to the next.
 - Read stack traces completely
 - Note line numbers, file paths, error codes
 
-**Action:** Use `read_file` on the relevant source files. Use `search_files` to find the error string in the codebase.
+**Action:** Use `Read` on the relevant source files. Use `Grep` to find the error string in the codebase.
 
 ### 2. Reproduce Consistently
 
@@ -70,7 +70,7 @@ You MUST complete each phase before proceeding to the next.
 - Does it happen every time?
 - If not reproducible → gather more data, don't guess
 
-**Action:** Use the `terminal` tool to run the failing test or trigger the bug:
+**Action:** Use the `Bash` tool to run the failing test or trigger the bug:
 
 ```bash
 # Run specific failing test
@@ -124,14 +124,14 @@ THEN investigate that specific component.
 - Keep tracing upstream until you find the source
 - Fix at the source, not at the symptom
 
-**Action:** Use `search_files` to trace references:
+**Action:** Use `Grep` to trace references:
 
 ```python
 # Find where the function is called
-search_files("function_name(", path="src/", file_glob="*.py")
+Grep("function_name(", path="src/", file_glob="*.py")
 
 # Find where the variable is set
-search_files("variable_name\\s*=", path="src/", file_glob="*.py")
+Grep("variable_name\\s*=", path="src/", file_glob="*.py")
 ```
 
 ### Phase 1 Completion Checklist
@@ -156,10 +156,10 @@ search_files("variable_name\\s*=", path="src/", file_glob="*.py")
 - Locate similar working code in the same codebase
 - What works that's similar to what's broken?
 
-**Action:** Use `search_files` to find comparable patterns:
+**Action:** Use `Grep` to find comparable patterns:
 
 ```python
-search_files("similar_pattern", path="src/", file_glob="*.py")
+Grep("similar_pattern", path="src/", file_glob="*.py")
 ```
 
 ### 2. Compare Against References
@@ -314,17 +314,17 @@ If you catch yourself thinking:
 
 Use these tools during Phase 1:
 
-- **`search_files`** — Find error strings, trace function calls, locate patterns
-- **`read_file`** — Read source code with line numbers for precise analysis
-- **`terminal`** — Run tests, check git history, reproduce bugs
-- **`web_search`/`web_extract`** — Research error messages, library docs
+- **`Grep`** — Find error strings, trace function calls, locate patterns
+- **`Read`** — Read source code with line numbers for precise analysis
+- **`Bash`** — Run tests, check git history, reproduce bugs
+- **`WebSearch`/`WebFetch`** — Research error messages, library docs
 
-### With delegate_task
+### With Agent
 
 For complex multi-component debugging, dispatch investigation subagents:
 
 ```python
-delegate_task(
+Agent(
     goal="Investigate why [specific test/behavior] fails",
     context="""
     Follow systematic-debugging skill:
@@ -337,7 +337,6 @@ delegate_task(
     File: [path to failing code]
     Test command: [exact command]
     """,
-    toolsets=['terminal', 'file']
 )
 ```
 

package/.agent/skills/test-driven-development/SKILL.md

@@ -277,25 +277,25 @@ Can't check all boxes? You skipped TDD. Start over.
 
 ### Running Tests
 
-Use the `terminal` tool to run tests at each step:
+Use the `Bash` tool to run tests at each step:
 
 ```python
 # RED — verify failure
-terminal("pytest tests/test_feature.py::test_name -v")
+Bash("pytest tests/test_feature.py::test_name -v")
 
 # GREEN — verify pass
-terminal("pytest tests/test_feature.py::test_name -v")
+Bash("pytest tests/test_feature.py::test_name -v")
 
 # Full suite — verify no regressions
-terminal("pytest tests/ -q")
+Bash("pytest tests/ -q")
 ```
 
-### With delegate_task
+### With Agent
 
 When dispatching subagents for implementation, enforce TDD in the goal:
 
 ```python
-delegate_task(
+Agent(
     goal="Implement [feature] using strict TDD",
     context="""
     Follow test-driven-development skill:
@@ -309,7 +309,6 @@ delegate_task(
     Project test command: pytest tests/ -q
     Project structure: [describe relevant files]
     """,
-    toolsets=['terminal', 'file']
 )
 ```
 

package/.agent/skills/video-orchestrator/SKILL.md

@@ -146,7 +146,7 @@ task graphs. See **[references/examples.md](references/examples.md)**.
 
 3. **One workspace per project.** All profiles for a given video share the same
    `dir:` workspace. Tasks pass artifacts via shared filesystem and structured
-   handoffs. **Every** `kanban_create` call passes
+   handoffs. **Every** `TaskCreate` call passes
    `workspace_kind="dir"` + `workspace_path="<absolute project path>"`.
 
 4. **Tenant every project.** Use a project-specific tenant

package/.agent/skills/web-pentest/SKILL.md

@@ -16,11 +16,11 @@ triggers:
   - "find vulns in [URL]"
   - "OWASP test [URL]"
 toolsets:
-  - terminal
-  - web
-  - browser
-  - file
-  - delegation
+  - Bash
+  - WebFetch
+  - Read
+  - Write
+  - Agent
 ---
 
 # Web Application Penetration Testing
@@ -200,11 +200,11 @@ input vectors.
 
 ## Phase 3: Vulnerability Analysis
 
-One delegate_task per vulnerability class. Each agent reads
+One Agent per vulnerability class. Each agent reads
 `evidence/recon.md` (+ `evidence/pre-recon.md` if present), produces
 `findings/<class>-queue.json` using `templates/exploitation-queue.json`.
 
-Use `delegate_task` with these focused subagents (parallel where possible):
+Use `Agent` with these focused subagents (parallel where possible):
 
 | Class | Goal | Reference |
 |-------|------|-----------|

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/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-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/wf-orchestrate-review.md

@@ -0,0 +1,35 @@
+---
+description: "4-lens parallel review — factual, domain, safety, style — coordinated as one panel and merged into a single verdict"
+---
+# /mindforge:wf-orchestrate-review
+
+Runs the **Orchestrate Review** dynamic workflow.
+
+## Usage
+`/mindforge:wf-orchestrate-review [target artifact, or describe what to review]`
+
+## What it does
+- **Scope**: Pins the target and any domain context
+- **Panel**: 4 parallel lenses — factual accuracy, domain correctness, safety, style/voice
+- **Consensus**: Merges all findings, deduplicates, assigns severity (blocking/major/minor)
+- **Verdict**: APPROVED / APPROVED_WITH_SUGGESTIONS / CHANGES_REQUIRED / BLOCKING
+
+## Running
+
+Invoke via Claude Code's Workflow tool:
+
+```
+Workflow({
+  scriptPath: ".mindforge/dynamic-workflows/scripts/orchestrate-review.js",
+  args: "<your input>"
+})
+```
+
+`args` can be a plain string target, or `{ target, domainContext }` for reviews that need
+subject-matter framing (e.g. "this is a medical dosage guide" so the domain lens knows
+what expertise to check against).
+
+Or discover via CLI:
+```bash
+node bin/mindforge-cli.js workflow info orchestrate-review
+```

package/.claude/commands/mindforge/wf-verification-loop.md

@@ -0,0 +1,45 @@
+---
+description: "Implementer vs. scorer loop: revise each round, score against a fixed metric, keep re-attempting until the score is maxed or a round cap is hit"
+---
+# /mindforge:wf-verification-loop
+
+Runs the **Verification Loop** dynamic workflow.
+
+## Usage
+`/mindforge:wf-verification-loop [task description, or { task, metric, targetScore, maxRounds }]`
+
+## What it does
+- **Scope**: Pins the artifact, the metric, and the target score / max-round cap
+- **Round**: An implementer agent revises the artifact; a separate scorer agent grades it
+  against the same fixed rubric every round; repeats until the target score is hit, the
+  score plateaus for `plateauRounds` consecutive rounds, or `maxRounds` is reached
+- **Report**: Round-by-round score history plus the final artifact
+
+Unlike a fixed-count review pass, this loop's whole point is that the implementer keeps
+re-attempting the SAME task, judged by a separate scorer, until the score stops moving —
+not a one-shot audit.
+
+## Running
+
+Invoke via Claude Code's Workflow tool:
+
+```
+Workflow({
+  scriptPath: ".mindforge/dynamic-workflows/scripts/verification-loop.js",
+  args: {
+    task: "Tighten this system prompt for token efficiency without losing instruction coverage",
+    metric: "10 pts clarity, 10 pts token economy, 10 pts no lost instructions",
+    targetScore: 95,
+    maxRounds: 8
+  }
+})
+```
+
+`args` can also be a plain string task description if you want the scorer to infer and
+hold its own rubric. Defaults: `targetScore` 95, `maxRounds` 8, `plateauRounds` 2 (stop
+early if the score doesn't improve for 2 consecutive rounds).
+
+Or discover via CLI:
+```bash
+node bin/mindforge-cli.js workflow info verification-loop
+```

package/.claude/commands/mindforge/wf-workflow-optimizer.md

@@ -0,0 +1,51 @@
+---
+description: "Builder + scorer + a distinct process-optimizer role: proposes structural changes to the PROCESS itself across rounds, not just fixes to one output"
+---
+# /mindforge:wf-workflow-optimizer
+
+Runs the **Workflow Optimizer** dynamic workflow.
+
+## Usage
+`/mindforge:wf-workflow-optimizer [task family description, or { taskFamily, metric, maxRounds, priorHistory }]`
+
+## What it does
+- **Scope**: Pins the task family, the fixed outcome metric, and any prior round history
+  you supply (this script holds no state of its own between invocations)
+- **Cycle**: A builder agent runs under the *current* process instructions; a separate
+  scorer agent grades the output against the fixed metric; a **third, distinct
+  process-optimizer agent** — with no role in building or scoring — diagnoses *why* the
+  process produced that result and rewrites the builder's instructions for next round.
+  Repeats until the optimizer judges the process has converged or `maxRounds` is hit.
+- **Report**: The score trend across rounds alongside every round's structural diagnosis,
+  so you can see whether rising scores came from real process changes, not just luck.
+
+This is not "implementer vs. scorer, keep re-attempting the same task" (that's
+`/mindforge:wf-verification-loop`, one artifact, one metric, no separate optimizer role).
+This workflow's object of improvement is the **process itself** — the builder's own
+instructions evolve round to round, and the deliverable includes that evolution history,
+not just a final artifact.
+
+## Running
+
+Invoke via Claude Code's Workflow tool:
+
+```
+Workflow({
+  scriptPath: ".mindforge/dynamic-workflows/scripts/workflow-optimizer.js",
+  args: {
+    taskFamily: "Write a one-paragraph executive summary of a technical incident postmortem",
+    metric: "30 pts non-technical-reader clarity, 30 pts root-cause traceability, 40 pts action-item concreteness",
+    maxRounds: 5
+  }
+})
+```
+
+To continue optimizing the SAME process across separate sessions, pass the previous run's
+returned `processHistory` back in as `args.priorHistory` — the script itself has no
+filesystem access and keeps no state between invocations, so cross-session continuity is
+the caller's responsibility.
+
+Or discover via CLI:
+```bash
+node bin/mindforge-cli.js workflow info workflow-optimizer
+```

package/.mindforge/config.json

@@ -1,11 +1,11 @@
 {
-  "version": "11.8.3",
+  "version": "11.9.1",
   "environment": "development",
   "governance": {
     "drift_threshold": 0.75,
     "critical_drift_threshold": 0.5,
     "res_threshold": 0.8,
-    "active_did": "did:mindforge:46cb14a1-0490-458e-9afb-947bf7cfba2b"
+    "active_did": "did:mindforge:ea8ace55-c2f0-46f2-af28-4d5a8ee3293a"
   },
   "revops": {
     "market_registry": {
@@ -71,7 +71,7 @@
   },
   "mesh": {
     "_node_id_note": "Set node_id to a unique value per node in multi-node deployments. auto = hostname-derived.",
-    "node_id": "auto",
+    "node_id": "beta-node",
     "peers": []
   },
   "ase": {

package/.mindforge/dynamic-workflows/REGISTRY.md

@@ -1,6 +1,6 @@
 # MindForge Dynamic Workflow Registry
 
-32 pre-built multi-agent workflows across 5 tiers. Each runs via Claude Code's `Workflow` tool using `parallel()`, `pipeline()`, `phase()`, and `agent()` primitives to fan out work across concurrent subagents and synthesize results.
+35 pre-built multi-agent workflows across 5 tiers. Each runs via Claude Code's `Workflow` tool using `parallel()`, `pipeline()`, `phase()`, and `agent()` primitives to fan out work across concurrent subagents and synthesize results.
 
 **Trigger:** Use `/mindforge:wf-catalog` to browse interactively, or invoke any `/mindforge:wf-<name>` command directly.
 
@@ -28,6 +28,8 @@
 | debug-detective | /mindforge:wf-debug-detective | 4-hypothesis parallel investigation → scientific RCA | Hard-to-reproduce bugs |
 | writer-reviewer | /mindforge:wf-writer-reviewer | Implement → fresh context review → verdict | Unbiased code review |
 | mutation-testing | /mindforge:wf-mutation-testing | Mutant generator → parallel kill-test → score | Test suite effectiveness |
+| orchestrate-review | /mindforge:wf-orchestrate-review | 4-lens parallel review: factual/domain/safety/style → verdict | Reviewing content/decisions, not code — docs, policies, plans |
+| verification-loop | /mindforge:wf-verification-loop | Implementer vs. scorer, iterate rounds until score maxed | Pushing a single metric as high as possible, not a fixed-pass audit |
 | code-explainer | /mindforge:wf-code-explainer | Structure → domain → architecture → narrative tour | Onboarding to unfamiliar codebases |
 | design-system-audit | /mindforge:wf-design-system-audit | 5-dimension parallel audit → consistency score | Design system health checks |
 
@@ -52,6 +54,7 @@
 | documentation-gen | /mindforge:wf-documentation-gen | Parallel doc gen → normalize → publish-ready | Generating or refreshing docs |
 | api-migration | /mindforge:wf-api-migration | Breaking change detection → guide → compat matrix | API versioning and migration |
 | data-pipeline-validate | /mindforge:wf-data-pipeline-validate | Stage-by-stage validation → quality gates | Data pipeline correctness checks |
+| workflow-optimizer | /mindforge:wf-workflow-optimizer | Builder → scorer → distinct process-optimizer, rounds evolve the process itself | Improving a repeatable process across attempts, not one output |
 
 ## Beast tier — Compound workflows (5 phases, 8+ agents, adversarial verification)