tml

Detect and fix documentation violations across C++ compiler and TML library code. Scans for missing ///, undocumented public items, wrong comment styles, and auto-fixes using parallel agents. Use when the user says "fix docs", "audit docs", "documenta", "corrige docs", or wants to improve documentation coverage.

Deep knowledge of TML compiler internals. Inject before working on compiler C++ code. Maps the full pipeline with key functions, types, and cross-subsystem boundaries.

Deep knowledge of TML standard library structure, module dependencies, key types, and C runtime mapping. Inject before working on library code.

Detect which tests are affected by recent code changes. Use when the user wants to know what tests to run after making changes.

List build artifacts with sizes and ages. Use when the user asks about build outputs, compiled binaries, or disk usage.

Build the TML compiler from C++ sources. Use when the user says "build", "compila", "rebuild", or needs to recompile the compiler after C++ changes.

Detect what changed and build only the necessary target. Use when the user says "build smart", "build inteligente", or wants an optimized build.

Invalidate compilation cache for specific source files. Use when cached results seem stale or after external changes to source files.

Type check a TML source file without compiling. Use when the user wants to verify types, check for errors, or validate code without building.

Stage and commit changes with a well-crafted conventional commit message. Use when the user says "commit", "comite", or asks to save changes.

Full Rust-as-Reference IR comparison workflow. Write equivalent .rs/.tml files, compile both, compare IR side-by-side. Use when the user says "compare ir", "compara ir", "rust reference", or needs to debug codegen quality.

Compile a TML source file to executable. Use when the user wants to compile a .tml file.

Show modules with lowest test coverage and suggest what to test next. Use when the user says "coverage gaps", "lacunas", "what needs tests", or wants to prioritize testing.

Emit LLVM IR for a TML source file. Use when the user wants to see the generated IR, debug codegen, or compare IR output.

Emit MIR (Mid-level IR) for a TML source file. Use for debugging the MIR pipeline.

Explain a TML compiler error code. Use when the user asks about an error code like T001, B001, L003.

Fix a codegen bug in the TML compiler. Use when the user says "fix codegen", "corrige codegen", "codegen bug", or when a test fails due to incorrect LLVM IR generation.

Format TML source files. Use when the user says "format", "formata", or wants to auto-format code.

Investigate a failing test by running it, emitting IR, and analyzing the error. Use when the user says "investigate", "investiga", "debug test", or wants to understand why a test fails.

Lint TML source files for style and potential issues. Use when the user says "lint" or wants to check code quality.

List all available test suites with file and test counts. Use when user asks "what suites exist", "list suites", "quais suites", or needs to know available test modules.

Analyze a C runtime file and suggest pure TML replacement. Use when the user says "migrate", "migra", "replace C", or wants to convert C runtime code to TML.

Optimize LLVM IR quality for a specific codegen pattern. Use when the user says "optimize ir", "otimiza ir", "improve codegen", or wants to make the compiler generate better LLVM IR.

Run format, lint, and affected tests before committing. Use when the user says "precommit", "pre-commit", "antes de comitar", or wants to validate changes before commit.

Run a quality audit on the codebase or a specific module. Use when the user says "qa", "quality", "audit", "qualidade", or wants to assess code quality, find issues, or create improvement tasks.

Clean build of the TML compiler. Use when the user says "rebuild", "clean build", "build limpo", or needs a fresh compilation.

Review a pull request or the current branch changes. Use when the user says "review", "review pr", "revise", or wants code review feedback.

Build and execute a TML source file. Use when the user says "run", "roda", "execute", or wants to run a .tml file.

Analyze which tests are slowest and why. Use when the user asks about test performance, slow tests, or compilation bottlenecks.

Project health dashboard showing test count, coverage, active tasks, and build status. Use when the user says "status", "dashboard", "estado do projeto", or wants a project overview.

Show the TML project module tree with file counts and test coverage. Use when the user asks about project structure, module layout, or file organization.

Archive a completed Rulebook task. Use when the user says "archive task", "arquiva task", or wants to move a completed task to the archive.

Create a new Rulebook task with proper template. Use when the user says "create task", "cria task", "nova task", or wants to start a new tracked work item.

List active Rulebook tasks with status and completion. Use when the user says "tasks", "tarefas", "list tasks", or wants to see what's in progress.

Test-driven development loop. Run a test, show errors, fix, repeat until passing. Use when the user says "tdd", "test driven", or wants an iterative test-fix cycle.

Run TML tests using the MCP test tool. Use when the user says "test", "roda os testes", "run tests", or wants to verify changes. Supports running specific test files, filtered tests, specific suites, or the full suite.

Run tests for a specific suite/module only. Use when the user says "testa core/str", "test std/json", "roda os testes de str", or wants to verify a specific module. Much faster than full suite.

Build the compiler and run targeted tests to verify changes. Use when the user says "verify", "verifica", "build and test", or after making compiler C++ changes that need validation.

Write new TML tests for a module or feature. Use when the user says "write tests", "escreve testes", "add tests", "test coverage for X", or wants to improve test coverage for a specific module.