Renamed three implementation workflows to avoid confusion:
- batch-super-dev → batch-stories (processes multiple stories)
- super-dev-pipeline → story-full-pipeline (single story with validation/review)
- story-pipeline → story-dev-only (basic TDD implementation only)
Changes include:
- Renamed workflow directories
- Updated workflow.yaml name fields and installed_path references
- Updated all cross-references in workflow.md files
- Updated state file naming conventions for clarity
- Updated references in documentation and other workflows
This naming convention eliminates pattern-matching confusion by using
completely distinct names for each workflow level.
**Slash Command Detection Fix:**
- Check for <command-name> tag explicitly (not pseudocode)
- If tag exists: Run in main context (interactive)
- If no tag: Delegate to Task agent (internal call)
- Reduces task nesting from 3 levels to 1 level
**Resilience Documentation:**
- Added RESILIENCE-FIX.md documenting state tracking
- Proposes state file for resume capability
- File read retry logic
- Error handling improvements
**Expected improvement:**
- Slash commands: No Task wrapper (1 level nesting)
- Fewer file access issues
- Can resume after crashes
To be implemented in future release.
**Problem:**
- workflow.xml enforced Task agent for ALL workflow executions
- User slash commands like /batch-super-dev need interactive prompts
- Can't run in Task agent (no user interaction)
**Solution:**
- Detect invocation source
- User slash command: Allow in main context
- Internal workflow call: Force Task agent delegation
**Detection:**
- Slash command: User message contains '/bmad_' or <command-name> tag
- Internal call: Invoked from within another workflow step
**Benefits:**
- Interactive workflows work (user can respond to prompts)
- Internal calls still get fresh context (Task agents)
- Balance between enforcement and UX
Tested: /bmad_bmm_batch-super-dev now runs without violation error.
**Enforcement Fixes from craftedcall (commits 9e299817e through 63e719d77):**
1. **workflow.xml** - Mandatory Task agent delegation
- FORBIDS executing workflows in main context
- REQUIRES workflow-executor Task agent
- Prevents context bloat and degradation
2. **step-enforcement.xml** (NEW)
- Centralized enforcement rules
- Task checkbox verification
- Dev Agent Record requirements
3. **workflow-executor.md** (NEW)
- Agent dedicated to workflow execution
- Loads ALL context upfront
- Executes with fresh context
4. **ALL-STEPS-EMBEDDED.md** (NEW)
- All 12 steps in single file
- Prevents agents from skipping steps
- Complete pipeline visibility
5. **step-04-implement.md** - Per-task checkbox enforcement
- Verify after EVERY task
- Auto-fix with 3-attempt retry
- Batch task verification
6. **step-10-complete.md** - Pre-commit verification
- Verify checked tasks > 0
- Auto-populate Dev Agent Record
- HALT only after auto-fix exhausted
7. **step-01-init.md** - Canonical filename enforcement
- ONE format only
- Auto-rename legacy files
8. **multi-agent-review/workflow.yaml** - Canonical format
These fixes were battle-tested on Epic 18 and proven to work.
All 352 unit tests passing.
Agents marked stories 'done' but left ALL tasks unchecked, Dev Agent Records empty,
and implemented different code than story specifications. This happened because
workflows had NO ENFORCEMENT - everything was optional theater.
**Enforcement Added (7 files changed):**
1. **workflow.xml** - Filename validation at completion
- Verifies output matches canonical format
- Auto-renames wrong-named files
- Prevents 'story-' prefix proliferation
2. **step-04-implement.md** - Per-task checkbox enforcement
- After EVERY task: Verify checkbox updated
- Auto-fix with Edit tool if missing
- 3-attempt retry before halt
- Batch tasks: Loop through all, verify each
3. **step-10-complete.md** - Final verification before commit
- Count checked tasks (HALT if zero)
- Verify Dev Agent Record filled (HALT if empty)
- Auto-fix from git commit if missing
4. **batch-super-dev Step 4.5** - Mandatory reconciliation
- Verify checked_tasks > 0 after reconciliation
- Auto-fix: Read commit → match tasks → check boxes
- Fill Dev Agent Record from commit message
- Override status to 'in-progress' if <80% complete
5. **super-dev-pipeline/workflow.yaml** - Canonical filename pattern
- Changed: story-*.md → *.md (removed 'story-' prefix)
6. **multi-agent-review/workflow.yaml** - Canonical format
- Changed: story-{story_id}.md → {story_id}.md
7. **batch-super-dev instructions** - ONE canonical format
- Format: {epic}-{story}-{slug}.md (NO 'story-' prefix)
- Auto-rename legacy 'story-' files
- Removed all 6 pattern variations
- Single format everywhere
**Behavioral Changes:**
- Before: Agents could skip tasks, leave checkboxes unchecked, claim 'done'
- After: Auto-fix with retry, HALT only if all attempts exhausted
- Before: Multiple filename formats caused confusion
- After: ONE format, auto-rename legacy files
- Before: No verification, trust agent claims
- After: Verify EVERYTHING, auto-fix failures
**Tested in production:**
- Epic 18 batch execution revealed all failure modes
- Every enforcement added addresses real observed failure
- No theoretical fixes - all battle-tested
**Quality Standards:**
Lives are at stake. Zero tolerance for unchecked tasks or missing documentation.
Ref: craftedcall commits 9e299817e, e607a4422, 6a9d87176, 0240b4742, 91f73c7f4
When provided, the style_guide input overrides all generic principles
(including Microsoft Style Guide baseline, reader-type priorities, and
structure-model selection) except CONTENT IS SACROSANCT.
Changes to both editorial-review-structure.xml and editorial-review-prose.xml:
- Add style_guide input after content input
- Add STYLE GUIDE OVERRIDE instruction in llm section
- Add "Consult style_guide" action in Step 3 for mid-flow refresh
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Co-authored-by: Brian <bmadcode@gmail.com>
Replace ambiguous "execute" terminology with explicit "Read fully and follow:"
phrasing across all workflow files to prevent LLM goal-seeking behavior where
models attempt to "achieve the end result" rather than following step-by-step
instructions verbatim.
Changes:
- Update 5 handler templates with canonical phrasing
- Replace ~150 INSTRUCTIONAL patterns across 87 workflow files
- Add "[Workflow] complete." prefix to 7 workflow endpoints
- Preserve BEHAVIORAL/STRUCTURAL patterns (agent descriptions, XML tags)
- Fix gitignore and markdownlint to ignore all node_modules directories
Closes#1372
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Add an optional also_consider parameter that allows callers to pass
domain-specific areas to keep in mind during review. This gently nudges
the reviewer toward specific concerns without overriding normal analysis.
Testing showed:
- Specific items steer strongly (questions get directly answered)
- Domain-focused items shift the lens (e.g., security focus = deeper security findings)
- Vague items have minimal effect (similar to baseline)
- Single items nudge without dominating
- Contradictory items handled gracefully
Includes test cases with sample content and 10 configurations to validate
the parameter behavior across different use cases.
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Co-authored-by: Brian <bmadcode@gmail.com>
* feat: add editorial review tasks for structure and prose
Add two complementary editorial review tasks:
- editorial-review-structure.xml: Structural editor that proposes cuts,
reorganization, and simplification. Includes 5 document archetype models
(Tutorial, Reference, Explanation, Prompt, Strategic) for targeted evaluation.
- editorial-review-prose.xml: Clinical copy-editor for prose improvements
using Microsoft Writing Style Guide as baseline.
Both tasks support humans and llm target audiences with different principles.
* fix: add content-sacrosanct guardrail to editorial review tasks
Both editorial review tasks (prose and structure) were missing the key
constraint that reviewers should never challenge the ideas/knowledge
themselves—only how clearly they are communicated. This restores the
original design intent.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* fix: align reader_type parameter naming across editorial tasks
Prose task was using 'target_audience' for the humans/llm optimization
flag while structure task correctly separates 'target_audience' (who
reads) from 'reader_type' (optimization mode). Aligns to reader_type.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Co-authored-by: Brian <bmadcode@gmail.com>
* fix(brainstorming): extend ideation phase with 100+ idea goal
Add emphasis on quantity-first approach to unlock better quality ideas.
Introduce energy checkpoints, multiple continuation options, and clearer
success metrics to keep users in generative exploration mode longer.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* fix(brainstorming): improve exploration menus and fix workflow paths
- Change menu options from numbers [1-4] to letters [K/T/A/B/C] for clearer navigation
- Fix workflow references from .yaml to .md across agents and patterns
- Add universal facilitation rules emphasizing 100+ idea quantity goal
- Update exploration menu with Keep/Try/Advanced/Break/Continue options
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* feat(brainstorming): implement research-backed procedural rigor
Phase 4 achievements:
- Added Anti-Bias Protocol (Every 10 ideas domain pivot)
- Added Chain-of-Thought requirements (Reasoning before generation)
- Implemented Simulated Temperature prompts for higher divergence
- Standardized Idea Format Template for quality control
- Fixed undefined Advanced Elicitation variables
- Synchronized documentation with new [K, T, A, P, C] pattern
* fix(brainstorming): align ideation goals and fix broken workflow paths
- Standardized quantity goals to 100+ ideas across all brainstorming steps
- Fixed broken .yaml references pointing to renamed .md workflow files
- Aligned documentation summaries with mandatory IDEA FORMAT TEMPLATE
- Cleaned up misplaced mindset/goal sections in core workflow file
- Fixed spelling and inconsistencies in facilitation rules
* Fix ambiguous variable names in brainstorming ideation step
* fix(brainstorming): enforce quality growth alongside quantity
* fix: correct dependency format and add missing frontmatter variable
---------
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
* fix: restore agent files accidentally modified in Docusaurus merge
Restores 25 agent files to their pre-merge state:
- Trigger format with shortcuts (WS, CH, BP, etc.) restored
- Fuzzy matching syntax restored
- BMB module: restores separate agent-builder, module-builder,
workflow-builder agents; removes consolidated bmad-builder
Also updates test to match restored trigger format.
Note: Schema validation needs update in follow-up commit.
* fix: normalize trigger fuzzy match format for schema validation
- Add dashes to fuzzy match text to match kebab-case triggers
- Add missing 'chat' kebab in CH triggers (CH or chat or fuzzy match on chat)
- Relax schema to allow 1-3 char shortcuts and skip shortcut derivation check
- Remove compound-wrong-shortcut test fixture (no longer validated)
All 24 agent files now pass schema validation.
* feat: add documentation website with Docusaurus build pipeline
* feat(docs): add AI discovery meta tags for llms.txt files
- Add global headTags with ai-terms, llms, llms-full meta tags
- Update landing page link to clarify AI context purpose
* fix(docs): restore accidentally deleted faq.md and glossary.md
Files were removed in 12dd97fe during path restructuring.
* fix(docs): update broken project-readme links to GitHub URL
* feat(schema): add compound trigger format validation
- Add markdownlint-cli2 as dev dependency
- Add lint:md script to package.json
- Add markdownlint job to CI workflow
- Configure 5 rules: heading-increment, no-duplicate-heading,
no-trailing-punctuation, no-bare-urls, no-space-in-emphasis
- Fix existing violations across 19 markdown files
- No auto-fix to prevent destructive changes
Closes#1034🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Brian <bmadcode@gmail.com>
* update 2024 to 2025
* fix(workflows): remove hardcoded years from WebSearch queries
Years in search queries (2024/2025) do not improve results - search
engines already prioritize current documentation. Tested all patterns
and confirmed identical quality results with/without years.
Removes years from:
- step-03-starter.md (5 queries)
- step-04-decisions.md (2 queries)
- game-architecture/instructions.md (2 queries)
Leaves file-utils.md unchanged (test fixture data, not a search query).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix(workflows): remove year placeholders from research WebSearch queries
Search engines return current results regardless of year - removes
{{current_year}} and hardcoded 2025 from step-05-technical-trends.md
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* refactor(workflows): replace {{current_year}} with semantic alternatives
Replaces year placeholder with context-appropriate wording:
- 'current data' for up-to-date information
- 'web searches' without year qualifier
- Updated failure mode to focus on using web searches
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix(workflows): clarify failure mode about stale training data
Rephrased to explicitly mention training data cutoff as the reason
to use web searches for current technology trends.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* refactor(workflows): make web search references platform-agnostic
- Remove hardcoded year references from WebSearch queries
- Replace `WebSearch` tool name with natural language "search the web"
- Soften "training data is stale" to "verify and supplement your knowledge"
- Add web search prerequisite check to research workflow
- Add platform-agnostic design note to CLAUDE.md
This framework targets 15+ agentic platforms, not just Claude Code.
Tool-specific syntax like `WebSearch:` won't work across all platforms.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* refactor(research): clean up prompts and routing
---------
Co-authored-by: Pomazan Bohdan <pomazan.bogdan@gmail.com>
Co-authored-by: Brian <bmadcode@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
Jonah Schulte
Alex Verkhovsky
Brian Madison
Eduard Voiculescu
Davor Racic