- Eager todo enforcement now skips prompts ending with question marks or exclamation marks, treating them as queries or commands rather than statements requiring task planning.
- Added 2 test cases validating that eager todo enforcement is skipped for prompts ending with question and exclamation marks.
- Added fully-qualified anchor paths in chunk read output showing complete path hierarchy with CRC checksums.
- Added exact chunk selector resolution with CRC validation to enforce canonical chunk paths in edit operations.
- Added `?` selector option to read syntax-aware chunks by canonical path listing when anchor paths are unclear.
- Improved tree-sitter range handling with `end_row_as_line()` function to correctly convert half-open ranges to 1-indexed line numbers.
- Extracted anchor label logic into `chunk_anchor_label()` const function and consolidated style-based selection.
- Removed --dev flag support from native build pipeline and dev:native npm scripts.
- Fixed plan review previews to re-append at the chat tail on refresh, keeping them adjacent to the active selector instead of updating off-screen.
- Added comprehensive test coverage for plan review rendering behavior across multiple refresh cycles.
- Refactored chunk selector resolution to extract and preserve raw selector strings separately from parsed CRC checksums.
- Reorganized chunk state lookup infrastructure with dedicated maps for checksum, leaf, and suffix-based chunk resolution.
- Simplified batch operation CRC validation to auto-accept stale checksums for subsequent operations on the same chunk.
- Unified TypeScript interface and Nix attrset chunk classification with consistent naming and nested expression handling.
- Extracted clipboard utility functions to dedicated module with improved OSC 52 and Termux support.
- Updated chunk read formatting with anchorStyle parameter and ellipsis standardization for improved consistency.
- Replaced nested operation objects with a flat edit schema using a single `target` field.
- Merged `sel` and `crc` into a unified `target: "chunk#CRC"` string copied from read output.
- Replaced `operations` array with `edits` and added `append`, `prepend`, `after`, `before` fields for insert positioning.
- Removed `defaultSelector` and `defaultCrc` top-level params; CRC is now always explicit per-edit.
- Migrated native bindings from TypeScript wrappers to NAPI-RS generated modules with auto-generated type definitions and runtime enums.
- Replaced chunk tree API with stateful ChunkState class supporting render, edit, and resolve operations with improved error handling.
- Converted callback signatures to error-first pattern (error, result) for shell, PTY, glob, and grep operations.
- Introduced type-safe enums for MacOSAppearance, GrepOutputMode, KeyEventType, ImageFormat, and AstMatchStrictness replacing string literals.
- Refactored chunk tree implementation with dedicated modules for edit, indent, resolve, and state management with comprehensive validation.
- Moved clipboard utilities from native bindings to coding-agent package with improved OSC 52 and Termux compatibility.
- Added `anchorStyle` setting ("full", "kind", "bare") to control anchor name verbosity in chunk output.
- Changed anchor tag delimiters from `` to `[…]` and gutter separator from `│` to `|`.
- Changed edit response to render only changed hunks with context instead of the full chunk tree.
- Fixed zero-width insert detection to derive target indent from neighboring lines instead of replaced range.
- Improved indent selection for body-level insertions by picking deeper indent between two neighbors.
- Updated error messages to clarify insert vs splice operations and line range validation.
- Modified assertChecksum() return type to handle cases with default CRC and no explicit selector.
- Distributed language-specific AST classification logic from defaults module to individual language modules for 15+ languages.
- Simplified defaults module to minimal catch-all fallbacks, removing 341 lines of match-based classification logic.
- Consolidated chunk edit operations by removing splice operation and unifying with line-scoped replace parameters.
- Simplified chunk rendering output format by reducing thresholds and changing line count display from 'lines' to 'ln' suffix.
- Removed PI_CHUNK_SPLICES environment variable and chunkSplicesEnabled conditional logic throughout codebase.
- Changed chunk edit operation format from flat `op` field to nested keyed union structure with operation name as object key.
- Added parseKeyedChunkOp() function to normalize both keyed and legacy flat operation formats for backward compatibility.
- Enhanced chunk selector sanitization with filename prefix stripping and case-insensitive CRC checksum matching.
- Updated chunk edit prompt documentation and test suite to reflect new nested operation JSON structure.
- Added `new_segment` parameter to `init_experiment` to force new segment creation when contract fields match.
- Added `skip_restore` parameter to `log_experiment` to preserve working tree state and pre-existing uncommitted changes.
- Changed `log_experiment` to revert only run-modified files instead of entire working tree, preserving user changes.
- Changed `init_experiment` to detect matching contract fields and skip re-initialization unless `new_segment=true`.
- Added git status parsing utilities to track pre-run dirty paths and distinguish tracked vs untracked modifications.
- Removed secondary metrics validation requirement, making secondary metrics informational only.
- Fixed indentation and formatting across multiple files for consistency.
- Reformatted code blocks to use tabs instead of spaces and improved line breaks for readability.
- Added connection timeout configuration to benchmark runner for early abort on no events.
- Implemented two-phase timeout strategy in benchmark runner with connection and activity phases.
- Added LSP diagnostic versioning support to track document versions and suppress stale diagnostics.
- Added options parameter to `waitForDiagnostics` and `getDiagnosticsForFile` with version filtering capabilities.
- Enabled `versionSupport` in LSP client capabilities to receive diagnostic version information from servers.
- Added test coverage for stale diagnostic suppression scenarios in LSP diagnostic freshness.
- Fixed memory leak by cancelling idle compaction timer on event controller disposal.
- Fixed session resumption to preserve last non-empty session when starting fresh.
- Fixed stash detection to use git ref resolution instead of output parsing for reliability.
- Fixed secret obfuscation to deobfuscate restored session messages locally while keeping LLM messages obfuscated.
- Fixed stash pop operation to preserve staged changes with --index flag after task branch merges.
- Changed idle compaction settings from enum to numeric type for flexible configuration.
- Fixed PR checkout tool to resolve worktree paths to canonical form using fs.realpath().
- Refactored test mocks to use vi.spyOn for git module functions instead of inline implementations.
- Updated test setup to pass Settings.isolated() with edit.manageImports enabled to EditTool.
- Added afterEach hooks across test suites to restore mocks after each test execution.
Wire EventBus through task runtime so subagent lifecycle and progress
events propagate to the TUI. Add SessionObserverRegistry to track
active sessions, and a session observer overlay accessible via Ctrl+S
that shows a picker of running subagents and a read-only transcript
viewer that reads the subagent's session JSONL file to display
thinking, text, tool calls, and results.
- Add TASK_SUBAGENT_LIFECYCLE_CHANNEL for start/end events
- Add SubagentProgressPayload.sessionFile for session file tracking
- Pass eventBus through sdk.ts -> main.ts -> InteractiveMode
- SessionObserverRegistry with multi-listener onChange pattern
- Overlay picker preserves selection position across live refreshes
- Viewer renders full transcript from session JSONL (thinking, text,
tool calls with smart arg summaries, inline tool results)
When entering plan mode, the thinking level configured on the plan role
(e.g., 'anthropic/claude-sonnet-4-5:xhigh') was discarded because
#resolveRoleModel() called resolveModelRoleValue() but only returned
.model, dropping the thinking level.
This is the same class of bug that was fixed in cycleRoleModels (which
correctly preserves and applies thinking level from role configs).
#applyPlanModeModel was the remaining unfixed call site.
Changes:
- Refactor #resolveRoleModel to #resolveRoleModelFull returning the
full ResolvedModelRoleValue instead of just Model
- Add resolveRoleModelWithThinking() public method exposing the full
resolution result
- Add optional thinkingLevel param to setModelTemporary() for atomic
model+thinking switches (eliminates two-step set-then-apply pattern)
- Bundle model+thinking into paired state objects in InteractiveMode
(#planModePreviousModelState, #pendingModelSwitch) — eliminates
4 parallel fields that could desync
- Simplify #applyPlanModeModel, flushPendingModelSwitch, #exitPlanMode
to single setModelTemporary calls
- Collapse cycleRoleModels two-step into single setModelTemporary call
- Add 6 tests for resolveRoleModelWithThinking
- Added language-specific import management for TypeScript, JavaScript, Python, Rust, Go, and C/C++ with automatic deduplication and grouping.
- Added `imports` parameter to patch, hashline, and replace edit operations for structured import/include requests with named, default, namespace, and alias support.
- Added `edit.manageImports` setting to control post-edit import management behavior across all edit operations.
- Implemented language-specific import handlers with parsing, merging, and rendering logic for each supported language.
- Integrated import application into EditTool with feature flag support and comprehensive test coverage across all languages.
- Migrated GitHub CLI utilities from gh-cli.ts module to utils/git.ts github namespace.
- Consolidated GitHub CLI command execution (available, run, json, text) into git.github namespace with unified error handling.
- Removed gh-cli.ts module and ControlledGit class infrastructure; updated all tool imports and call sites.
- Exported git utilities from main package entry point for extension use.
- Extracted git operations from ControlledGit class into centralized utils/git module with 1276 lines of typed command wrappers.
- Replaced dependency injection of ControlledGit instances with direct cwd string parameters across commit agent tools and workflows.
- Migrated all git command execution from inline shell calls and custom helpers to structured git module API (diff, status, branch, worktree, patch, etc.).
- Removed ControlledGit class, operations.ts, and helper functions (findGitHeadPath, mergeStdoutStderr, joinPatch) now provided by git module.
- Exported git utilities from main package entry point for extension and plugin use.
- Consolidated fetch tool into read tool with URL reading capability and caching support.
- Removed standalone fetch tool from all agent prompts and CLI documentation.
- Extended read tool schema with timeout and raw parameters for URL fetch control.
- Added URL caching mechanism to prevent redundant network requests during read operations.
- Refactored fetch module from class-based tool to standalone executeReadUrl function.
- Updated read tool documentation to describe multi-purpose capabilities including web pages, GitHub, Stack Overflow, Wikipedia, Reddit, NPM, arXiv, blogs, and feeds.
- Added support for writing to ZIP archives using fflate library for cross-platform compatibility.
- Modified archive writing to detect and handle ZIP files separately from Bun archives.
- Removed GhPrPushTool test case.
- Added support for writing to archive entries using `archive.ext:path/inside/archive` syntax for .tar, .tar.gz, .tgz, and .zip formats.
- Added ability to create new archives automatically when writing to non-existent archive subpaths with nested directory support.
- Implemented archive path resolution and validation with fallback to plain file writes for non-archive filenames.
- Added comprehensive test coverage for archive write operations including entry updates, archive creation, and fallback behavior.
- Removed 3 test cases from gh.test.ts that relied on implementation details and external state.
- Removed unused helper functions getCurrentHeadSha() and GhToolDetails import.
- Updated test assertions to use path.resolve() and corrected expected remote name from fork-contrib to forksrc.
- Reformatted else blocks in shell.rs for consistency.
- Replaced Python-based markitdown CLI with native markit-ai library for document and notebook conversion.
- Added support for converting Jupyter notebooks (.ipynb) to markdown via markit-ai integration.
- Created markit utility module with file/buffer conversion wrappers and standardized error handling.
- Removed markitdown from Python tools manager and eliminated external CLI dependency.
- Updated fetch, read, and scraper tools to use new markit conversion API across codebase.
- Added test coverage for ipynb file conversion through markit utility.
- Added archive reading support for .tar, .tar.gz, .tgz, and .zip files with virtual subpath syntax (archive.ext:path/inside/archive).
- Added archive listing and subdirectory navigation capabilities to explore archive contents without extraction.
- Added /tools slash command to display visible agent tools and their capabilities.
- Implemented ArchiveReader module with openArchive(), parseArchivePathCandidates(), and archive content extraction methods.
- Added fflate 0.8.2 dependency for archive decompression support.
- Added comprehensive test coverage for TAR, TAR.GZ, TGZ, and ZIP archive reading with fixture generation utilities.
- abortableSleep used Bun.sleep which vitest fake timers do not intercept;
switch to setTimeout via Promise.withResolvers so vi.useFakeTimers() +
vi.advanceTimersByTime() work correctly in the polling loop tests
- getCurrentHeadSha() and GhRunWatchTool session cwd used a hardcoded
/work/pi path; replace with import.meta.dir so the tests resolve the
real repo HEAD on any machine
- add timeout-minutes: 30 to the CI test job as a safety net against
future hangs blocking release runs
- Added gh-renderer.ts module with GitHub Actions workflow visualization for terminal output.
- Enhanced gh_pr_view tool to fetch and display inline review comments alongside pull request reviews.
- Improved gh_run_watch tool output with structured job state tracking and failed log formatting.
- Fixed gh_run_watch to resolve explicit branch watches against GitHub API branch head instead of local HEAD.
- Fixed gh_* tool outputs to spill full large results to artifacts instead of pre-truncating.
- Standardized trailing newlines across 100+ prompt template files for POSIX compliance.
- Added gh_pr_checkout tool to check out GitHub pull requests into git worktrees with branch metadata.
- Added gh_pr_push tool to push checked-out pull request branches to source fork with force-with-lease support.
- Implemented 15+ git helper functions for worktree, branch, and remote management operations.
- Extended GhPrViewData interface with headRefOid, headRepository, isCrossRepository, and maintainerCanModify fields.
- Added comprehensive integration tests for gh_pr_checkout and gh_pr_push tools with fixture setup and cleanup.
sanitizeRehydratedOpenAIResponsesAssistantMessage now removes the
assistant providerPayload for Responses-family messages, not just the
stale thinkingSignature. After rehydration the native replay snapshot
belongs to a previous live provider connection and replaying it on a
warmed session causes 401 rejections from GitHub Copilot.
User/developer providerPayload is preserved by the caller since it
carries durable compaction and preserved history needed for valid cold
request reconstruction.
Fixes#592
can1357