fix: pass empty file list directly instead of converting to None in code review agent

fix: deduplicate read_source_samples call and filter test files from code review

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

fix

fix

fix

fix: add missing file_list param to read_source_samples and wire changed_files through

fix: fix: remove unused python_files variable in code_review_agent after refactor

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

fixed lint

style(test_generation_agent): fix import order, remove trailing whitespace

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

style(repo_analysis_agent): fix extra indent on comment, add blank line before run

style(specialist_review_agent): fix import order, split long import lines

style(code_review_agent): fix import order, remove trailing whitespace, add blank line

fix(security): resolve all bandit findings — log swallowed exception, annotate justified subprocess usage

feat(report): include specialist security/perf/arch findings in final report

fix(report_generator): accept and forward specialist_review parameter

fix(agents): export specialist_review_agent from package __init__

fixed

fix

fix

feat(tools): add diff-aware file filtering for incremental reviews

feat(agents): add auto-fix suggestion generation to code_review_agent

feat(agents): add specialist multi-agent review with judge synthesis

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

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

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(test_generation_agent): strip markdown code fences from LLM response before storing test_code

fix(test_generation_agent): use pathlib.Path.name instead of str.rsplit to fix Windows path handling

chore(agents): explicitly export all agent modules from package __init__

fix(code_review_agent): replace broken manual JSON parse with call_llm_for_json; remove unused call_llm import

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

fix(repo_analysis_agent): remove dead re-parse block, llm_response was never assigned

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

fix(agents): replace direct model mutation with model_copy for Pydantic v2 correctness

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

fix(agents): replace pytest --co with coverage run to actually measure coverage

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

fix: check only filename not full path when skipping test files in code review agent

fix(agent): check filename only for test_ skip, not full path

fix: remove eager wildcard imports from __init__.py files

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