- Add ListRecipes and ListAvailablePrompts functions
- Update MCP server handlers for local/global recipes and prompts
- Add unit tests for analyze, docs, mcp, prompts, recipe, and testgen packages
- Implement `grokkit mcp` command to run as MCP server over stdio
- Export analysis, docs, and testgen functions for MCP tools
- Add tools: lint_code, analyze_code, generate_docs, generate_tests, generate_commit_msg, run_recipe
- Register resources: recipes://local, recipes://global, prompts://language
- Update README and add user guide for MCP
- Add MCP config options and dependencies
This change adds `_ =` assignments to exec.Command(...).Run() calls in test setup code to explicitly ignore error returns, likely to satisfy linters or static analysis tools without changing behavior.
- Add colorized output for success/error messages
- Change custom message flag from -m to -M
- Implement branch prefixing (feature/ or fix/)
- Add config for workon.model and workon.ide
- Update README.md index on task completion
- Integrate IDE opening if configured
- Refine error handling and logging
Replace placeholder with actual Grok client call to generate and append a tailored work plan section to todo/fix files. Add prompt engineering for concise, actionable plans. Include file content reading and model config integration. Update comments and error handling for clarity.
- Remove temporary grokClient interface and prompt logic.
- Add TODOs for real Grok client import and calls.
- Placeholder work plan for functionality.
- TODOs for README index update and IDE config support.
- Minor comment and logging tweaks for clarity.
- Bootstrap todo structure and handle fix/complete modes.
- Create safe git branches and append AI-generated work plans.
- Commit changes and move items to completed on finish.
- Add stubs for Grok client and config-dependent IDE open.
- Update workonCmd to accept todo_item_title arg and add -f/--fix, -c/--complete flags
- Implement transactional flow: bootstrap todo dirs, move/create .md files, create branch, append Grok-generated Work Plan, commit
- Add todo package with Bootstrap for directory structure
- Expand workon.Run to handle modes (todo, fix, complete) with placeholders for Grok integration and optional cnadd/IDE open
Introduces the `workon` CLI command which selects the next queued todo item,
moves it to doing/, creates a git branch, generates a work plan via Grok,
appends it to the file, and commits the changes. Includes skeleton implementation
with TODOs for full functionality.
- Introduce git.Diff function that uses CombinedOutput and treats exit code 1 as success (indicating changes exist).
- Update prdescribe command to use git.Diff and trim whitespace when checking for empty diffs.
- Enhance prdescribe output to include base branch in the status message.
Add a "none" linter configuration for the Rexx language to provide structure,
while noting that no traditional linter is used and it's primarily for grokkit analyze.
- Remove `rootCmd.AddCommand(analyzeCmd)` from `cmd/analyze.go` init func
- Add empty `Linters` array for Rexx in `internal/linter/linter.go` with comment for analyze support
- Change default output filename in analyze command from analyze.md to analysis.md
- Add Rexx language detection with extensions .rx, .rex, .rexlib, .rexx, .cls
Update DetectPrimaryLanguage to normalize "C/C++" and "C++" to "c" for consistent counting.
Revise function comment and internal comments for clarity.
Remove redundant comment on finding most common language.
- Fix config.GetModel to use command name and flag
- Switch to package-level logger functions
- Update git.IsRepo to take no arguments
- Simplify linter language detection comments
- Adjust Grok client creation to NewClient().StreamSilent
- Add error handling for confirmation input
- Remove unnecessary imports and refine comments in linter
- Updated Go analysis prompt for clarity, structure, and educational focus.
- Improved buildProjectContext to include shallow key files and cleaner Git remote handling.
- Implemented DetectPrimaryLanguage with counting logic and Go bias; added SupportedLanguages.
- Enhanced LoadAnalysisPrompt with better language handling, fallbacks, and error clarity.
- Add Go-specific analysis prompt in .grokkit/prompts/go.md
- Expand cmd/analyze.go to discover files, detect language, load prompts, build context, generate report via Grok, and handle output with preview/confirmation
- Integrate analyzeCmd into root command
- Introduce internal/linter/language.go for primary language detection
- Add internal/prompts/analyze.go for loading analysis prompts from project or global locations
- Introduce new test suite in internal/recipe/recipe_test.go covering recipe loading, parameter overrides, safety checks, work dir resolution, file discovery, and unified patch creation.
- Remove t.Parallel() from tests in cmd/changelog_test.go, cmd/root_test.go, and cmd/scaffold_test.go that modify global state (e.g., os.Args, HOME, or use os.Chdir()) to avoid race conditions and ensure test isolation.
- Introduce global safeCommands() map for command whitelisting.
- Implement case-insensitive prefix checking for allowed commands.
- Simplify argument handling by removing redundant int conversions.
- Update error messages and comments for clarity on security policies.
- Remove outdated comments and adjust prompt text for consistency.
Update the command matching logic to require an exact match or the command followed by a space and arguments. Also normalize case and trim whitespace for safe commands to prevent loose prefix matches that could allow unintended commands.
Update the executeReadOnlyShell function to handle numbers in command arguments,
such as 'tree -L 3', by changing Args to []interface{} and converting them to strings
before execution. Add strconv import for formatting.
- Tighten trigger conditions for read-only shell steps to specific phrases
- Add robust JSON extraction with escaped quote handling
- Reorder user confirmation before whitelist check in execution flow
- Relocate FileChange struct and clean up comments
- Update recipe markdown for step title consistency
- Introduce new Step 1 in result-refactor.md for exploring project structure using safe read-only shell commands (e.g., tree, cat).
- Rename subsequent steps accordingly.
- Add handler in runner.go for read-only shell steps triggered by keywords like "explore" or "inspect".
Implement executeReadOnlyShell method to safely run whitelisted read-only commands,
prompting for user approval and integrating AI-suggested commands for filesystem context.
- Replace hardcoded safeCommands map with sync.OnceValue loading from ~/.config/grokkit/safe_shell_commands.yaml
- Provide fallback built-in safe list if config load fails
- Add safe_shell_commands.yaml.example for user reference
- Update safety check to use loaded map and prefix matching
- Renamed safeReadOnlyCommands to safeCommands for clarity.
- Added support for additional safe commands including GNU utilities (find, grep, which),
Git commands (diff, branch), and various test runners (go test, make test/lint, pytest, etc.).
- Updated safety check to allow commands prefixed with any safe command for flexibility.
- Improved error message for unsafe commands.
Update the error message for unsafe shell commands to be more concise and user-friendly, removing redundant "ERROR:" prefix and "try again" instruction.
- Consolidate resolveWorkDir and remove resolvePackagePath for cleaner path handling.
- Eliminate executeShellCommands and related logic to disable shell execution in recipes.
- Simplify error messaging in loader for unsafe commands.
Implement a read-only command whitelist in the recipe loader to reject
potentially dangerous shell commands, ensuring only safe operations like
ls, pwd, cat, etc., are permitted. This enhances security by preventing
execution of unauthorized commands in recipes.
- Rename and expand resolvePackagePath to resolveWorkDir, which now appends project_name if provided
- Update discoverFiles to use workDir and add default search pattern if none specified
- Modify executeShellCommands prompt to include workDir and enforce relative paths only
- Add safety checks to reject shell commands with ".." or absolute paths to prevent directory escapes
- Minor prompt and formatting adjustments for clarity and consistency
- Introduce resolvePackagePath to handle ~ expansion, relative paths, and absolutization at the start.
- Update discoverFiles and executeShellCommands to use the resolved workDir.
- Remove redundant path logic from discoverFiles and shell execution for better maintainability.
Introduce a new mechanism in the recipe runner to execute whitelisted shell commands for steps like initialization, creation, or running tools (e.g., poetry, git). Commands are generated via AI prompt, parsed from JSON, validated against an allowed list, and executed within a strict working directory to ensure safety.
- Add SearchPattern field to Recipe struct
- Update discoverFiles to use configurable search pattern, defaulting to "if err != nil"
- Set default search_pattern in result-refactor.md recipe
Update discoverFiles to leverage recipe metadata for extensions and apply smart defaults more cleanly. Generalize logic with comments for future improvements, while retaining err != nil check for now.
Enhance discoverFiles to respect explicit package_path param if provided.
If no param is given, intelligently default to 'src' directory if it exists,
otherwise fall back to project root (.).
Introduce ResolvedParams field to Recipe struct for storing resolved
parameter values from defaults and user overrides. Update loader to
populate and use it for template rendering. Adjust runner to use
ResolvedParams for root path and generalize file discovery.
Introduce ProjectLanguages and Extensions fields to the Recipe struct
to support option 2 for generic file discovery. Also update comments
for internal fields populated by the loader.
- Build allowed extensions from recipe languages and extensions map
- Generalize filepath walking to filter by allowed extensions instead of hardcoding .go
- Retain Go-specific content check for now; can generalize later
- Simplify refactor JSON handling and patch creation by removing redundant comments and error checks
- Adjust comments and minor formatting for clarity
- Introduce refactorJSONs to collect pure JSON from refactor steps for apply.
- Update discoverFiles to respect recipe's package_path parameter.
- Refine handleApplyStep to parse and apply changes more robustly.
- Remove outdated test output files.
Add refactorFiles to process discovered files individually, generating small JSON responses per file to avoid truncation. Update handleApplyStep to collect and parse multiple single-file JSONs into a unified patch. Switch discoverFiles comment to reflect real scanning. Add fallback default in Run for other steps.
- Blank import bufio to avoid unused warnings.
- Refine regex to handle double-quoted format and fix backtick issues.
- Update comments for clarity on regex changes.
Adjust the regex in handleApplyStep to flexibly match both old and new Grok formats for code blocks. Also, remove blank import for bufio as it's now used.
- Update discoverFiles comment to be more concise.
- Enhance blockRe regex to optionally match leading "// " for filenames, supporting varied Grok output formats.
- Revise handleApplyStep comment to reflect regex changes.
- Simplify createUnifiedPatch by removing unnecessary error checks on fmt.Fprintf and defer closure.
- Adjust bufio import to blank (likely for side effects or pending use)
- Refine comments for clarity and remove unnecessary ones
- Split regex string to avoid backtick collisions in literals
- Add error handling to patch writing in createUnifiedPatch
- Minor formatting and defer close adjustments
