style: fix isort and ruff issues in CLI files and tests

Author: jeremyeder

.claude/commands/review-agentready.md

@@ -0,0 +1,192 @@
+---
+allowed-tools: Bash(gh issue view:*), Bash(gh search:*), Bash(gh issue list:*), Bash(gh pr comment:*), Bash(gh pr diff:*), Bash(gh pr view:*), Bash(gh pr list:*), Bash(git:*), Bash(python:*)
+description: AgentReady-specific code review with attribute mapping and score impact analysis
+disable-model-invocation: false
+---
+
+Provide an AgentReady-specific code review for the given pull request.
+
+This command extends the standard `/code-review` with agentready project-specific concerns:
+- Map findings to the 25 agentready attributes (from agent-ready-codebase-attributes.md)
+- Calculate impact on self-assessment score (current: 80.0/100 Gold)
+- Generate remediation commands (agentready bootstrap, black, pytest, etc.)
+- Link to relevant CLAUDE.md sections
+- Categorize by severity with confidence scoring
+
+## Process
+
+Follow these steps precisely:
+
+1. **Eligibility Check** (Haiku agent)
+   - Check if PR is closed, draft, or already reviewed
+   - Skip if automated PR (dependabot, renovate) unless it touches assessors
+   - If ineligible, exit early
+
+2. **Context Gathering** (Haiku agent)
+   - List all CLAUDE.md files (root + modified directories)
+   - Get full PR diff and metadata
+   - Return concise summary of changes
+
+3. **Parallel AgentReady-Focused Review** (5 Sonnet agents)
+   Launch 5 parallel agents to independently review:
+
+   **Agent #1: CLAUDE.md Compliance Audit**
+   - Check adherence to CLAUDE.md development workflows
+   - Verify pre-push linting (black, isort, ruff)
+   - Verify test requirements (pytest, >80% coverage)
+   - Check branch verification, conventional commits
+
+   **Agent #2: AgentReady-Specific Bug Scan**
+   Focus on agentready assessment logic:
+   - TOCTOU bugs in file system operations
+   - AST parsing correctness (false positives/negatives in assessors)
+   - Measurement accuracy issues
+   - Type annotation correctness
+   - Error handling patterns (try-except, graceful degradation)
+
+   **Agent #3: Historical Context Analysis**
+   - Read git blame for modified assessor files
+   - Check for regression in assessment accuracy
+   - Verify attribute scoring logic hasn't changed unintentionally
+
+   **Agent #4: Previous PR Comment Analysis**
+   - Review comments on past PRs touching same files
+   - Check for recurring issues
+
+   **Agent #5: Code Comment Compliance**
+   - Verify changes follow inline comment guidance
+
+4. **Attribute Mapping** (Haiku agent for each issue)
+   For each issue found:
+   - Map to specific agentready attribute ID (e.g., "2.3 Type Annotations")
+   - Determine tier (1=Essential, 2=Critical, 3=Important, 4=Advanced)
+   - Calculate score impact using tier weights (Tier 1: 50%, Tier 2: 30%, etc.)
+   - Generate remediation command (black, pytest, agentready bootstrap --fix)
+   - Link to CLAUDE.md section
+
+5. **Confidence Scoring** (parallel Haiku agents)
+   For each issue, score 0-100 confidence:
+   - 0: False positive
+   - 25: Might be real, unverified
+   - 50: Verified but minor
+   - 75: Very likely real, important
+   - 90: Critical issue (auto-fix candidate)
+   - 100: Blocker (definitely auto-fix)
+
+   **Critical Issue Criteria** (confidence ≥90):
+   - Security vulnerabilities (path traversal, injection)
+   - TOCTOU race conditions
+   - Assessment accuracy bugs (false positives/negatives)
+   - Type safety violations causing runtime errors
+   - Missing error handling leading to crashes
+
+6. **Filter Issues**
+   - Keep issues with confidence ≥80 for reporting
+   - Flag issues with confidence ≥90 as "auto-fix candidates"
+   - Calculate aggregate score impact
+
+7. **Final Eligibility Check** (Haiku agent)
+   - Verify PR is still eligible for review
+   - Check if PR was updated during review
+
+8. **Post Review Comment** (using gh pr comment)
+   Use the custom AgentReady format (see below)
+
+## AgentReady Review Output Format
+
+Use this format precisely:
+
+---
+
+### 🤖 AgentReady Code Review
+
+**PR Status**: [X issues found] ([Y 🔴 Critical], [Z 🟡 Major], [W 🔵 Minor])
+**Score Impact**: Current 80.0/100 → [calculated score] if all issues fixed
+**Certification**: Gold → [Platinum/Gold/Silver/Bronze] potential
+
+---
+
+#### 🔴 Critical Issues (Confidence ≥90) - Auto-Fix Recommended
+
+##### 1. [Brief description]
+**Attribute**: [ID Name] (Tier [N]) - [Link to CLAUDE.md section]
+**Confidence**: [90-100]%
+**Score Impact**: [−X.X points]
+**Location**: [GitHub permalink with full SHA]
+
+**Issue Details**:
+[Concise explanation with code snippet if relevant]
+
+**Remediation**:
+```bash
+# Automated fix available via:
+# (Will be applied automatically if this is a blocker/critical)
+[specific command: black file.py, pytest tests/test_foo.py, etc.]
+```
+
+---
+
+#### 🟡 Major Issues (Confidence 80-89) - Manual Review Required
+
+##### 2. [Brief description]
+**Attribute**: [ID Name] (Tier [N])
+**Confidence**: [80-89]%
+**Score Impact**: [−X.X points]
+**Location**: [GitHub permalink]
+
+[Details and remediation as above]
+
+---
+
+#### Summary
+
+- **Auto-Fix Candidates**: [N critical issues] flagged for automatic resolution
+- **Manual Review**: [M major issues] require human judgment
+- **Total Score Improvement Potential**: +[X.X points] if all issues addressed
+- **AgentReady Assessment**: Run `agentready assess .` after fixes to verify score
+
+---
+
+🤖 Generated with [Claude Code](https://claude.ai/code)
+
+<sub>- If this review was useful, react with 👍. Otherwise, react with 👎.</sub>
+
+---
+
+## False Positive Examples
+
+Avoid flagging:
+- Pre-existing issues (not in PR diff)
+- Intentional functionality changes
+- Issues caught by CI (linters, tests, type checkers)
+- Pedantic style issues not in CLAUDE.md
+- General code quality unless explicitly in CLAUDE.md
+- Issues with lint-ignore comments
+- Code on unmodified lines
+
+## AgentReady-Specific Focus Areas
+
+**High Priority**:
+1. File system race conditions (TOCTOU in scanning logic)
+2. Assessment accuracy (false positives/negatives)
+3. Type annotations (Python 3.11+ compatibility)
+4. Error handling (graceful degradation)
+5. Test coverage (>80% for new code)
+
+**Medium Priority**:
+6. CLAUDE.md workflow compliance
+7. Conventional commit format
+8. Documentation updates (docstrings, CLAUDE.md)
+
+**Low Priority**:
+9. Performance optimization
+10. Code organization/structure
+
+## Notes
+
+- Use `gh` CLI for all GitHub operations
+- Always link to code with full SHA (not `HEAD` or branch names)
+- Link format: `https://github.com/owner/repo/blob/[full-sha]/path#L[start]-L[end]`
+- Provide ≥1 line context before/after issue location
+- Make todo list first
+- Calculate score impact using tier-based weighting from agentready scoring algorithm