跳轉至

Appendix A: Complete MCP Tool Reference

Total Tools: 100+ tools organized into 20+ categories. Version: V14.0.0 (Predictive Intelligence & Offline-First)

💡 Recommendation: Use the Universal Router (boring()) or CLI (boring-route) for most tasks. You rarely need to call these tools directly.


0. Universal Tools (The Gateway)

Tool Description Key Parameters
boring Universal Router: Natural Language interface to all tools. request (Natural Language)
boring_help Show all available categories and profile info. -
boring_discover Progressive Disclosure: Get full schema for a tool. tool_name

1. Verification Tools

Tool Description Key Parameters
boring_verify Run full project verification level: BASIC/STANDARD/FULL/SEMANTIC
boring_verify_file Verify a single file file_path, level

2. Security Tools

Tool Description Key Parameters
boring_security_scan SAST + secret detection + dependency scan project_path, scan_type, verbosity

Scan Types: secrets, vulnerabilities, dependencies, all Verbosity: minimal, standard, verbose


3. Transaction Tools

Tool Description Key Parameters
boring_transaction_start Create savepoint message
boring_transaction_commit Finalize transaction -
boring_rollback Restore to savepoint transaction_id

4. Background Task Tools

Tool Description Key Parameters
boring_background_task Submit async task task_type, task_args
boring_task_status Check task progress task_id
boring_list_tasks List all tasks status filter

Task Types: verify, test, lint, security_scan


5. Context & Memory Tools

Tool Description Key Parameters
boring_save_context Save session state context_name, data
boring_load_context Restore session state context_name
boring_list_contexts List saved contexts -
boring_get_profile Get user preferences -
boring_learn_fix Learn fix pattern error_pattern, fix_pattern

6. RAG (Semantic Search) Tools

Tool Description Key Parameters
boring_rag_index Index codebase force, project_path (Requires chromadb)
boring_rag_search Semantic code search query, max_results, verbosity
boring_rag_context Get code context file_path, function_name
boring_rag_expand Expand dependency graph chunk_id, depth
boring_rag_graph Visual Graph: Visualize code dependency graph. target_path, depth, output_format
boring_rag_status Check index health -

7. Multi-Agent Tools

Tool Description Key Parameters
boring_multi_agent [PROMPT GENERATOR / EXECUTOR] task, auto_approve_plans, execute
boring_agent_plan [PROMPT GENERATOR] Architect task
boring_agent_review [PROMPT GENERATOR] Reviewer code_change
boring_delegate Delegate to sub-agent task, tool_type

Tool Types: database, web_search, file_system, api, reasoning


8. Shadow Mode Tools

Tool Description Key Parameters
boring_shadow_mode Set mode mode: DISABLED/ENABLED/STRICT
boring_shadow_status View pending ops -
boring_shadow_approve Approve operation operation_id, note
boring_shadow_reject Reject operation operation_id, note
boring_shadow_clear Clear all pending -

9. Git & Hooks Tools

Tool Description Key Parameters
boring_hooks_install Install pre-commit hooks -
boring_commit Semantic commit message, files
boring_visualize Visualize git history -

10. Plugin Tools

Tool Description Key Parameters
boring_list_plugins List all plugins -
boring_run_plugin Execute plugin name, args
boring_reload_plugins Reload from disk -

11. Workspace Tools

Tool Description Key Parameters
boring_workspace_add Register project name, path, tags
boring_workspace_remove Unregister project name
boring_workspace_list List projects tag filter
boring_workspace_switch Switch context name

12. Auto-Fix Tools

Tool Description Key Parameters
boring_auto_fix Automated fix loop max_iterations, verification_level
boring_suggest_next AI suggestions (Context-aware: Git, Task, Patterns) limit
boring_get_progress Check long-running task task_id


13. Vibe Coder Pro Tools

Tool Description Key Parameters
boring_test_gen Auto-Test Generator: Generates unit tests (Py/JS/TS). file_path, output_dir
boring_code_review AI Code Review: Quality analysis with Brain pattern matching. file_path, verbosity
boring_perf_tips Performance Analysis: Detects N+1, I/O in loops, etc. file_path, verbosity
boring_arch_check Architecture Audit: Generates Mermaid dependency graphs. target_path, output_format
boring_doc_gen Documentation Generator: Automated API reference generation. target_path
boring_vibe_check Vibe Score: Health check (Lint + Security + Doc) with fix prompt. target_path, verbosity
boring_impact_check Impact Analysis: Reverse dependency tracking for safe changes. target_path, max_depth

14. Intelligence Tools (V11.2)

Tool Description Key Parameters
boring_predict_errors Error Prediction: Forecasts bugs based on local history. file_path, limit
boring_health_score Execution Efficiency: Analyzes success rates & solving speed. project_path
boring_predict_impact Risk Prediction: Estimates risk level for planned changes. file_path, action
boring_risk_areas Heatmap: Identifies files prone to recurrent failures. limit
boring_cache_insights Cache Audit: Insights into AdaptiveCache performance. -
boring_set_session_context Boost Context: Manually bias RAG/Intelligence. keywords

15. Evaluation & LLM Judge

Tool Description Key Parameters
boring_evaluate LLM Judge: Automated code scoring (1-5 targets). target, level, rubric
boring_generate_rubric Rubric Factory: Creates detailed evaluation criteria. task, criteria
boring_bias_report Monitoring: Detects position/length bias in scoring. days
boring_evaluation_metrics System Health: Correlation with human judgments. -

16. Knowledge & Brain Tools

Tool Description Key Parameters
boring_learn Active Learning: Extracts patterns from session logs. topics
boring_brain_summary Brain Dashboard: High-level status of persistent knowledge. -
boring_brain_health Health Report: Pattern decay and usage stats. -
boring_global_export Share: Export high-value patterns to Global Brain. -
boring_global_import Learn: Import patterns from Global Brain. tags
boring_create_rubrics Create evaluation standards from project specs. -

17. SpecKit Tools (Specification-Driven Development)

Tool Description Key Parameters
boring_speckit_plan Plan: Generate implementation plan from plan.md. workflow_file
boring_speckit_tasks Breakdown: Create actionable task list. workflow_file
boring_speckit_analyze Analyze: Check spec consistency. workflow_file
boring_speckit_clarify Clarify: Identify ambiguities in spec. workflow_file
boring_speckit_checklist Verify: Generate verification checklist. workflow_file
boring_speckit_constitution Rules: Create project guiding principles. workflow_file

18. Skills & Discovery Tools

Tool Description Key Parameters
boring_skills_install Install: Install Agent Skill. name
boring_skills_list List: List Skills Catalog. platform
boring_skills_search Search: Search Skills Catalog. query

19. Predictive Intelligence Tools (V14.0)

These tools leverage historical data and Brain patterns to predict issues and optimize workflow.

Tool Description Key Parameters
boring_predict_impact Predict change impact: Analyzes risk level and affected files based on history. file_path, change_type
boring_risk_areas Identify hotspots: Shows files with high error frequency. limit
boring_cache_insights Cache analytics: Shows hit rates and correlation prefetch stats. -
boring_set_session_context Optimize Context: Tune RAG/Cache for specific task type (debug/feature). task_type, keywords
boring_get_session_context Check Context: View current optimization parameters. -

20. Supplemental Registered Tools (V14.0)

Tool Description Key Parameters
boring_active_skill Activate a skill category and inject tools skill_name
boring_best_next_action Suggest best next action based on project state -
boring_brain_status Brain status and context overview -
boring_brain_sync Sync global brain patterns -
boring_call Execute internal tool by name tool_name, arguments
boring_checkpoint Create or restore project checkpoint action, message
boring_context Manage context (save/load/list) action, context_id
boring_distill_skills Distill learned patterns into skills project_path
boring_done Completion notification message, title
boring_flow Run Flow engine workflow goal, mode
boring_get_relevant_patterns Get relevant brain patterns query, limit
boring_global_list List global patterns limit
boring_incremental_learn Incremental learning from changes project_path
boring_inspect_tool Inspect internal tool schema tool_name
boring_integrity_score Integrity score report project_path
boring_intelligence_stats Intelligence usage stats -
boring_learn_pattern Learn a single pattern pattern, solution
boring_optimize_context Optimize context window text, goal
boring_orchestrate Orchestrate multi-step workflow goal
boring_pattern_stats Brain pattern stats project_path
boring_profile Manage profile (get/learn) action, error_pattern
boring_prompt_fix Prompt generator for fixes task
boring_prompt_plan Prompt generator for planning task
boring_prune_patterns Prune low-quality patterns limit
boring_rag_reload Reload RAG dependencies -
boring_reset_skills Reset injected skills -
boring_session_auto Auto session workflow task
boring_session_confirm Confirm session step step_id
boring_session_load Load session session_id
boring_session_pause Pause session -
boring_session_start Start session goal
boring_session_status Session status -
boring_shadow_trust Trust path or command path
boring_shadow_trust_list List trust rules -
boring_shadow_trust_remove Remove trust rule path
boring_skill_activate Activate skill skill_name
boring_skill_create Create skill name, goal
boring_skill_discover Discover local skills project_path
boring_skill_download Download skill url
boring_skills_recommend Recommend skills project_path
boring_status Project status snapshot -
boring_synth_tool Synthesize new tool description
boring_task Background tasks (submit/status/list) action, task_type
boring_transaction Transactions (start/commit/rollback) action, description
boring_transaction_rollback Legacy rollback alias project_path
boring_usage_stats Usage statistics -

21. CLI Integration (V14.0)

Some V14.0 features are best accessed via CLI directly.

Command Description
boring predict Run predictive analysis on current changes.
boring bisect AI-powered git bisect to find bug sources.
boring diagnostic Deep project health check.
boring model Manage local LLMs (download/list/remove).
boring doctor System health and MCP connection check.
boring offline Toggle offline experience mode.

MCP Resources

Resource URI Description
boring://capabilities List all capability categories
boring://tools/{category} Detailed usage for category
boring://logs Server logs (stdio mode)

MCP Prompts

Prompt Description
plan_feature Generate implementation plan
review_code Request code review
debug_error Debug error message
refactor_code Request refactoring
explain_code Explain code
setup_project Project setup workflow
verify_work Verification workflow
manage_memory Memory management workflow
evaluate_architecture Hostile Architect review
run_agent Multi-agent workflow
vibe_start One-click Start - Full workflow (Plan → Build → Verify). Best for: "Build new app", "Start project"
quick_fix Quick Fix - Auto-fix lint/format/bugs. Best for: "Fix lint errors", "Clean up code"
full_stack_dev Full-stack development

Last updated: V14.0.0