chore: v2.2.1 — clean up gitignore, update docs

- Untrack marketing-diagram.excalidraw (already gitignored by pattern)
- Untrack evaluate/results/ and evaluate/reports/ (generated output)
- Update FEATURES.md, LLM-OPTIMIZED-REFERENCE.md, CHANGELOG.md for v2.2.1
- Bump version to 2.2.1

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: tirth8205

.gitignore

@@ -62,8 +62,10 @@ diagrams/*.excalidraw
 diagrams/export_pngs.mjs
 *.excalidraw
 
-# Evaluation test repos (third-party checkouts)
+# Evaluation (test repos + generated results/reports)
 evaluate/test_repos/
+evaluate/results/
+evaluate/reports/
 
 # Draft/duplicate files
 docs/assets/marketing-diagram*

CHANGELOG.md

@@ -1,5 +1,37 @@
 # Changelog
 
+## [2.2.1] - 2026-04-07
+
+### Added
+- **Parallel parsing**: `ProcessPoolExecutor` for 3-5x faster builds (`CRG_PARSE_WORKERS`, `CRG_SERIAL_PARSE`)
+- **Lazy post-processing**: `postprocess="full"|"minimal"|"none"` parameter, `run_postprocess` MCP tool + CLI command
+- **SQLite-native BFS**: Recursive CTE replaces NetworkX for impact analysis (`CRG_BFS_ENGINE`)
+- **Configurable limits**: `CRG_MAX_IMPACT_NODES`, `CRG_MAX_IMPACT_DEPTH`, `CRG_MAX_BFS_DEPTH`, `CRG_MAX_SEARCH_RESULTS`
+- **Multi-hop dependents**: N-hop `find_dependents()` with `CRG_DEPENDENT_HOPS` (default 2) and 500-file cap
+- **Token-efficient output**: `detail_level="minimal"` on 8 tools for 40-60% token reduction
+- **`get_minimal_context` tool**: Ultra-compact entry point (~100 tokens) with task-based tool routing
+- **Token-efficient prompts**: All 5 MCP prompts rewritten with minimal-first workflows
+- **Incremental flow/community updates**: `incremental_trace_flows()`, `incremental_detect_communities()`
+- **Visualization aggregation**: Community/file/auto modes with drill-down for large graphs (`--mode`)
+- **Token-efficiency benchmarks**: 5 workflow benchmarks in `eval/token_benchmark.py`
+- **DB schema v6**: Pre-computed `community_summaries`, `flow_snapshots`, `risk_index` tables
+- **Token Efficiency Rules** in all skill templates and CLAUDE.md
+
+### Changed
+- CLI `build`/`update` support `--skip-flows`, `--skip-postprocess` flags
+- PostToolUse hook uses `--skip-flows` for faster incremental updates
+- VS Code extension schema version bumped to v6
+
+### Fixed
+- mypy type errors in parallel parsing and context tool
+- Bandit false positive on prompt preamble string
+- Import sorting in graph.py, main.py, tools/__init__.py
+- Unused imports cleaned up in cli.py
+
+### Housekeeping
+- Gitignore: untrack `marketing-diagram.excalidraw`, `evaluate/results/`, `evaluate/reports/`
+- Updated FEATURES.md, LLM-OPTIMIZED-REFERENCE.md, CHANGELOG.md for v2.2.1
+
 ## [2.1.0] - 2026-04-03
 
 ### Added

docs/FEATURES.md

@@ -1,6 +1,21 @@
 # Features
 
-## v2.1.0 (Current)
+## v2.2.1 (Current)
+- **24 MCP tools** (up from 22): Added `get_minimal_context` and `run_postprocess`.
+- **Parallel parsing**: `ProcessPoolExecutor` for 3-5x faster builds on large repos.
+- **Lazy post-processing**: `postprocess="full"|"minimal"|"none"` to skip expensive steps.
+- **SQLite-native BFS**: Recursive CTE replaces NetworkX for impact analysis (faster on large graphs).
+- **Token-efficient output**: `detail_level="minimal"` on 8 tools for 40-60% token reduction.
+- **`get_minimal_context`**: Ultra-compact entry point (~100 tokens) with task-based tool routing.
+- **Incremental flow/community updates**: Only re-trace affected flows, skip community re-detection when unaffected.
+- **Visualization aggregation**: Community/file/auto modes with drill-down for 5k+ node graphs.
+- **Token-efficiency benchmarks**: 5 workflow benchmarks in eval framework.
+- **Pre-computed summary tables**: DB schema v6 with `community_summaries`, `flow_snapshots`, `risk_index`.
+- **Configurable limits**: `CRG_MAX_IMPACT_NODES`, `CRG_MAX_IMPACT_DEPTH`, `CRG_DEPENDENT_HOPS`, etc.
+- **Multi-hop dependents**: N-hop dependent discovery (default 2) with 500-file cap.
+- **615 tests** across 22 test files.
+
+## v2.1.0
 - **22 MCP tools** (up from 9): 13 new tools for flows, communities, architecture, refactoring, wiki, multi-repo, and risk-scored change detection.
 - **5 MCP prompts**: `review_changes`, `architecture_map`, `debug_issue`, `onboard_developer`, `pre_merge_check` workflow templates.
 - **18 languages** (up from 15): Added Dart, R, Perl support.

docs/LLM-OPTIMIZED-REFERENCE.md

@@ -1,4 +1,4 @@
-# LLM-OPTIMIZED REFERENCE -- code-review-graph v2.1.0
+# LLM-OPTIMIZED REFERENCE -- code-review-graph v2.2.1
 
 Claude Code: Read ONLY the exact `<section>` you need. Never load the whole file.
 
@@ -7,15 +7,15 @@ Quick install: pip install code-review-graph
 Then: code-review-graph install && code-review-graph build
 First run: /code-review-graph:build-graph
 After that use only delta/pr commands.
-For v2 features: detect_changes_tool for risk-scored reviews, list_flows_tool for execution paths, list_communities_tool for architecture.
+ALWAYS start with get_minimal_context_tool(task="your task") — returns ~100 tokens with risk, communities, flows, and suggested next tools.
+Use detail_level="minimal" on all subsequent calls unless you need more detail.
 </section>
 
 <section name="review-delta">
-Always call get_impact_radius on changed files first.
-Then get_review_context (depth=2).
-Or use detect_changes_tool for risk-scored, priority-ordered review guidance.
-Generate review using ONLY changed nodes + 2-hop neighbors.
-Target: <800 tokens total context.
+1. Call get_minimal_context_tool(task="review changes") first.
+2. If risk is low: detect_changes_tool(detail_level="minimal") → report summary.
+3. If risk is medium/high: detect_changes_tool(detail_level="standard") → expand on high-risk items.
+Target: ≤5 tool calls, ≤800 tokens total context.
 </section>
 
 <section name="review-pr">
@@ -24,10 +24,11 @@ Never include full files unless explicitly asked.
 </section>
 
 <section name="commands">
-MCP tools (22): build_or_update_graph_tool, get_impact_radius_tool, query_graph_tool, get_review_context_tool, semantic_search_nodes_tool, embed_graph_tool, list_graph_stats_tool, get_docs_section_tool, find_large_functions_tool, list_flows_tool, get_flow_tool, get_affected_flows_tool, list_communities_tool, get_community_tool, get_architecture_overview_tool, detect_changes_tool, refactor_tool, apply_refactor_tool, generate_wiki_tool, get_wiki_page_tool, list_repos_tool, cross_repo_search_tool
+MCP tools (24): get_minimal_context_tool, build_or_update_graph_tool, run_postprocess_tool, get_impact_radius_tool, query_graph_tool, get_review_context_tool, semantic_search_nodes_tool, embed_graph_tool, list_graph_stats_tool, get_docs_section_tool, find_large_functions_tool, list_flows_tool, get_flow_tool, get_affected_flows_tool, list_communities_tool, get_community_tool, get_architecture_overview_tool, detect_changes_tool, refactor_tool, apply_refactor_tool, generate_wiki_tool, get_wiki_page_tool, list_repos_tool, cross_repo_search_tool
 MCP prompts (5): review_changes, architecture_map, debug_issue, onboard_developer, pre_merge_check
 Skills: build-graph, review-delta, review-pr
-CLI: code-review-graph [install|init|build|update|status|watch|visualize|serve|wiki|detect-changes|register|unregister|repos|eval]
+CLI: code-review-graph [install|init|build|update|status|watch|visualize|serve|wiki|detect-changes|postprocess|register|unregister|repos|eval]
+Token efficiency: All tools support detail_level="minimal" for compact output. Always call get_minimal_context_tool first.
 </section>
 
 <section name="legal">
@@ -48,7 +49,7 @@ Configure via CRG_EMBEDDING_MODEL env var or model parameter.
 </section>
 
 <section name="languages">
-Supported (18): Python, TypeScript/TSX, JavaScript, Vue, Go, Rust, Java, Scala, C#, Ruby, Kotlin, Swift, PHP, Solidity, C/C++, Dart, R, Perl
+Supported (19): Python, TypeScript/TSX, JavaScript, Vue, Go, Rust, Java, Scala, C#, Ruby, Kotlin, Swift, PHP, Solidity, C/C++, Dart, R, Perl, Lua
 Parser: Tree-sitter via tree-sitter-language-pack
 </section>