feat: add security depth — secret scanning, dependency vulns, report surface

fix

fix: feat: fault-tolerant parallel agents, AST security scanner, and shared source reader

fixed lint

style(bug_detection_agent): remove trailing whitespace, add blank line before run

fix(imports): remove duplicate get_python_files and unused scan_repository in bug_detection_agent

security: apply sanitize_for_prompt before LLM submission in all three code-reading agents

style: fix indentation on comment lines in bug_detection_agent and repo_analysis_agent

fix(bug_detection_agent): remove dead llm_response re-parse block and fix indentation

fix(bug_detection_agent): remove dead llm_response re-parse block and fix indentation

feat(llm): add call_llm_for_json with corrective retry on invalid JSON response

feat(agents): add conditional LLM calls — skip when no signal, follow-up pass on low scores

fix: log warnings instead of silently swallowing file read errors in agents and scanner

fix: resolve all ruff lint errors (E402, F401, F811)

fix: move llm_data fallback inside except block in 3 agents

fix: strip markdown fences from LLM JSON responses before parsing

refactor: replace calculate_totals() with Pydantic model_validator

fix: surface LLM JSON parse failures with logging and fallback

refactor: replace print() with structured logging across all agents and tools

feat: convert to async/await with parallel agent execution via asyncio.gather

fix: suppress remaining bandit warnings with nosec comments

feat: add all 5 agents - analysis, bugs, tests, review, report