mirror of https://github.com/azaion/ai-training.git synced 2026-04-22 22:46:35 +00:00

Files

T

Oleksandr Bezdieniezhnykh cdcd1f6ea7 Fix .cursor skills consistency: flow resolution, tracker-agnostic refs, report naming, error recovery

- Rewrite autopilot flow resolution to 4 deterministic rules based on source code + docs + state file presence
- Replace all hard-coded Jira references with tracker-agnostic terminology across 30+ files
- Move project-management.mdc to _project.md (project-specific, not portable with .cursor)
- Rename FINAL_implementation_report.md to context-dependent names (implementation_report_tests/features/refactor)
- Remove "acknowledged tech debt" option from test-run — failing tests must be fixed or removed
- Add debug/error recovery protocol to protocols.md
- Align directory paths: metrics -> 06_metrics/, add 05_security/, reviews/, 02_task_plans/ to README
- Add missing skills (test-spec, test-run, new-task, ui-design) to README
- Use language-appropriate comment syntax for Arrange/Act/Assert in coderule + testing rules
- Copy updated coderule.mdc to parent suite/.cursor/rules/
- Raise max task complexity from 5 to 8 points in decompose
- Skip test-spec Phase 4 (script generation) during planning context
- Document per-batch vs post-implement test run as intentional
- Add skill-internal state cross-check rule to state.md

2026-03-28 02:34:00 +02:00

2.6 KiB

Raw Blame History

Phase 4: Execution

Role: Orchestrator Goal: Execute all refactoring tasks by delegating to the implement skill Constraints: No inline code changes — all implementation goes through the implement skill's batching and review pipeline

4a. Pre-Flight Checks

Verify refactoring task files exist in TASKS_DIR (created during Phase 2d):
- All [TRACKER-ID]_refactor_*.md files are present
- Each task file has valid header fields (Task, Name, Description, Complexity, Dependencies)
Verify TASKS_DIR/_dependencies_table.md includes the refactoring tasks
Verify all tests pass (safety net from Phase 3 is green)
If any check fails, go back to the relevant phase to fix

4b. Delegate to Implement Skill

Read and execute .cursor/skills/implement/SKILL.md.

The implement skill will:

Parse task files and dependency graph from TASKS_DIR
Detect already-completed tasks (skip non-refactoring tasks from prior workflow steps)
Compute execution batches for the refactoring tasks
Launch implementer subagents (up to 4 in parallel)
Run code review after each batch
Commit and push per batch
Update work item ticket status

Do NOT modify, skip, or abbreviate any part of the implement skill's workflow. The refactor skill is delegating execution, not optimizing it.

4c. Capture Results

After the implement skill completes:

Read batch reports from _docs/03_implementation/batch_*_report.md
Read the latest _docs/03_implementation/implementation_report_*.md file
Write RUN_DIR/execution_log.md summarizing:
- Total tasks executed
- Batches completed
- Code review verdicts per batch
- Files modified (aggregate list)
- Any blocked or failed tasks
- Links to batch reports

4d. Update Task Statuses

For each successfully completed refactoring task:

Transition the work item ticket status to Done via the configured tracker MCP
If tracker unavailable, note the pending status transitions in RUN_DIR/execution_log.md

For any failed or blocked tasks, leave their status as-is (the implement skill already set them to In Testing or blocked).

Self-verification:

All refactoring tasks show as completed in batch reports
All completed tasks have work item tracker status set to Done
All tests still pass after execution
No tasks remain in blocked or failed state (or user has acknowledged them)
RUN_DIR/execution_log.md written with links to batch reports

Save action: Write RUN_DIR/execution_log.md

GATE: All refactoring tasks must be implemented. If any tasks failed, present the failures to the user and ask for guidance before proceeding to Phase 5.

2.6 KiB Raw Blame History