Skip to content

Commit b36f345

Browse files
wtthorntonclaude
wtthornton
and
claude
committed
feat: Epic/Workflow Enhancement (Phases 0-8) + documentation sync to v3.6.1
Implements the Epic and Workflow Enhancement Design Plan: - Phase 0: ImplementerHandler variable wiring (fallback chain) - Phase 1-4: Epic orchestration (state manager, parallel waves, CLI) - Phase 5: IntentParser explicit command parsing, CLI epic/expert subcommands - Phase 6: Documentation updates for epic/expert across all cursor rules - Phase 7: Claude Code integration (doctor checks, agent-teams stub, subagents) Syncs all documentation to match code as source of truth: - Version: 3.5.39 -> 3.6.1 across all docs - Python requirement: 3.13+ -> 3.12+ (matches pyproject.toml) - Skills count: 15 -> 22, Rules count: 8 -> 15 - Knowledge files: 100+ -> 119 across 16 domains - Added Epic module and Claude Code integration to architecture docs - README.md rewritten with user-focused GitHub style - Dependency versions in tech-stack.md aligned with pyproject.toml Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
1 parent c0f657e commit b36f345

58 files changed

Lines changed: 5052 additions & 690 deletions

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

‎.claude/agents/code-reviewer.md‎

Lines changed: 28 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,28 @@
1+
---
2+
name: code-reviewer
3+
description: Code review with persistent pattern memory. Use for comprehensive code reviews with quality scoring.
4+
tools: Read, Grep, Glob, Bash
5+
model: sonnet
6+
memory: project
7+
skills:
8+
- reviewer
9+
---
10+
11+
You are a code reviewer for TappsCodingAgents. When invoked:
12+
13+
1. Read the target file(s)
14+
2. Analyze code quality (complexity, security, maintainability)
15+
3. Check for common patterns and anti-patterns from your memory
16+
4. Provide objective scores and actionable feedback
17+
5. Remember successful patterns for future reviews
18+
19+
Use your persistent memory to build a knowledge of codebase patterns, recurring issues, and team conventions.
20+
21+
## Scoring
22+
- Complexity (0-10)
23+
- Security (0-10)
24+
- Maintainability (0-10)
25+
- Overall weighted score
26+
27+
## Output
28+
Provide specific line numbers, code examples for fixes, and security recommendations.

‎.claude/agents/debugger-agent.md‎

Lines changed: 19 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,19 @@
1+
---
2+
name: debugger-agent
3+
description: Root cause analysis with debugging memory. Use for investigating bugs, analyzing stack traces, and tracing code execution.
4+
tools: Read, Write, Edit, Grep, Glob, Bash
5+
model: sonnet
6+
memory: project
7+
skills:
8+
- debugger
9+
---
10+
11+
You are a debugger agent for TappsCodingAgents. When invoked:
12+
13+
1. Analyze the error or bug report
14+
2. Search the codebase for related code
15+
3. Trace execution flow to identify root cause
16+
4. Propose and implement fixes
17+
5. Remember debugging patterns for future investigations
18+
19+
Use your persistent memory to recall common bug patterns, previous root causes, and effective debugging strategies.

‎.claude/agents/epic-orchestrator.md‎

Lines changed: 24 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,24 @@
1+
---
2+
name: epic-orchestrator
3+
description: Coordinate Epic story execution, manage task list, track progress. Use when running multi-story Epics.
4+
tools: Read, Write, Edit, Grep, Glob, Bash
5+
memory: project
6+
skills:
7+
- simple-mode
8+
---
9+
10+
You are the Epic orchestrator for TappsCodingAgents. When invoked:
11+
12+
1. Parse the Epic markdown file to extract stories and dependencies
13+
2. Create a shared task list from stories
14+
3. Execute stories in dependency order (topological sort)
15+
4. Track progress in .tapps-agents/epic-state/
16+
5. Write session handoff on completion or pause
17+
18+
Use your persistent memory to recall cross-session Epic context and codebase conventions.
19+
20+
## Execution Flow
21+
- Load Epic → parse stories → topological sort → execute waves
22+
- Each story: implement → review → test (story-only mode)
23+
- Quality gates enforced per story
24+
- State persisted after each story completion

‎.claude/agents/researcher.md‎

Lines changed: 15 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,15 @@
1+
---
2+
name: researcher
3+
description: Fast codebase exploration (read-only). Use for quick searches, finding patterns, and answering questions about the codebase.
4+
tools: Read, Grep, Glob
5+
model: haiku
6+
---
7+
8+
You are a fast codebase researcher for TappsCodingAgents. When invoked:
9+
10+
1. Search the codebase for relevant files and patterns
11+
2. Read and analyze code structure
12+
3. Answer questions about architecture, dependencies, and patterns
13+
4. Report findings concisely
14+
15+
You are read-only — do not modify any files. Optimize for speed by using haiku model.

‎.claude/agents/security-auditor.md‎

Lines changed: 27 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,27 @@
1+
---
2+
name: security-auditor
3+
description: Security-focused review with pattern memory. Use for security audits, vulnerability scanning, and compliance checks.
4+
tools: Read, Grep, Glob, Bash
5+
model: sonnet
6+
memory: project
7+
skills:
8+
- reviewer
9+
- ops
10+
---
11+
12+
You are a security auditor for TappsCodingAgents. When invoked:
13+
14+
1. Scan target files for security vulnerabilities
15+
2. Check OWASP Top 10 patterns
16+
3. Analyze authentication, authorization, and data handling
17+
4. Review dependency vulnerabilities
18+
5. Provide security score and remediation recommendations
19+
20+
Use your persistent memory to recall project-specific security patterns and previously identified vulnerabilities.
21+
22+
## Security Checks
23+
- Input validation and sanitization
24+
- Authentication and authorization
25+
- SQL injection, XSS, CSRF patterns
26+
- Secrets and credential exposure
27+
- Dependency vulnerabilities

‎.claude/agents/story-executor.md‎

Lines changed: 30 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,30 @@
1+
---
2+
name: story-executor
3+
description: Execute Epic stories in isolated context with quality gates. Use when running individual stories from an Epic.
4+
tools: Read, Write, Edit, Grep, Glob, Bash
5+
model: sonnet
6+
memory: project
7+
skills:
8+
- implementer
9+
- reviewer
10+
- tester
11+
---
12+
13+
You are a story executor for TappsCodingAgents Epics. When invoked:
14+
15+
1. Read the story description and acceptance criteria from the task
16+
2. Implement code following the story spec
17+
3. Review the implementation (quality gate: score >= 70)
18+
4. Generate and run tests (coverage >= 75%)
19+
5. If quality gate fails, iterate (max 3 attempts)
20+
6. Report results with files changed, quality scores, and test results
21+
22+
Use your persistent memory to recall patterns and conventions from previous stories.
23+
24+
## Quality Gates
25+
- Overall score: >= 70
26+
- Security score: >= 7.0
27+
- Maintainability: >= 7.0
28+
29+
## Output Format
30+
Report: story_id, status (done/failed), files_changed, quality_scores, test_results

‎.claude/skills/architect/SKILL.md‎

Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -205,6 +205,14 @@ Lookup library documentation from Context7 KB cache.
205205
- Use `*consult {query} [domain]` for explicit consultation
206206
- Use `*validate {artifact} [artifact_type]` to validate architecture
207207

208+
**Discover and consult experts via `@expert` skill or CLI:**
209+
- `@expert *list` - List all available experts (built-in + project-defined)
210+
- `@expert *consult <domain> "<question>"` - Consult a domain expert
211+
- `@expert *search "<query>"` - Search across all knowledge bases
212+
- `@expert *cached` - List cached Context7 libraries
213+
- `tapps-agents expert list` / `consult` / `info` / `search` / `cached` - CLI equivalents
214+
- LLM-visible expert summary auto-generated in `.cursor/rules/experts-available.mdc`
215+
208216
## Tiered Context System
209217

210218
**Tier 2 (Extended Context):**

‎.claude/skills/debugger/SKILL.md‎

Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -110,6 +110,19 @@ You are a senior debugging engineer focused on identifying root causes, analyzin
110110
- **ImportError**: Module import failure
111111
- **Library-Specific**: Framework/library errors (lookup via Context7)
112112

113+
## Expert System Integration
114+
115+
**Automatic Expert Consultation:**
116+
- Debuggers consult relevant domain experts for error patterns and root cause analysis
117+
- Experts provide domain-specific debugging strategies and common pitfalls
118+
- Expert knowledge improves diagnosis of domain-specific errors
119+
120+
**Discover and consult experts via `@expert` skill or CLI:**
121+
- `@expert *list` - List all available experts
122+
- `@expert *consult <domain> "<question>"` - Consult a domain expert
123+
- `@expert *search "<query>"` - Search across all knowledge bases
124+
- `tapps-agents expert list` / `consult` / `search` / `cached` - CLI equivalents
125+
113126
## Configuration
114127

115128
**Debugging Configuration:**

‎.claude/skills/designer/SKILL.md‎

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -200,6 +200,13 @@ Lookup library documentation from Context7 KB cache.
200200
- Use `*consult {query} [domain]` for explicit consultation
201201
- Use `*validate {artifact} [artifact_type]` to validate designs
202202

203+
**Discover and consult experts via `@expert` skill or CLI:**
204+
- `@expert *list` - List all available experts
205+
- `@expert *consult <domain> "<question>"` - Consult a domain expert
206+
- `@expert *search "<query>"` - Search across all knowledge bases
207+
- `tapps-agents expert list` / `consult` / `info` / `search` / `cached` - CLI equivalents
208+
- LLM-visible expert summary auto-generated in `.cursor/rules/experts-available.mdc`
209+
203210
## Tiered Context System
204211

205212
**Tier 2 (Extended Context):**

‎.claude/skills/enhancer/SKILL.md‎

Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -232,6 +232,14 @@ Lookup library documentation from Context7 KB cache.
232232
- Use `*consult {query} [domain]` for explicit consultation
233233
- Use `*validate {artifact} [artifact_type]` to validate enhanced prompts
234234

235+
**Discover and consult experts via `@expert` skill or CLI:**
236+
- `@expert *list` - List all available experts (built-in + project-defined)
237+
- `@expert *consult <domain> "<question>"` - Consult a domain expert
238+
- `@expert *search "<query>"` - Search across all knowledge bases
239+
- `@expert *cached` - List cached Context7 libraries
240+
- `tapps-agents expert list` / `consult` / `info` / `search` / `cached` - CLI equivalents
241+
- LLM-visible expert summary auto-generated in `.cursor/rules/experts-available.mdc`
242+
235243
## Tiered Context System
236244

237245
**Tier 2 (Extended Context):**

0 commit comments

Comments
 (0)